Impact of COVID-19 pandemic on the sleep quality of medical professionals in Brazil

ABSTRACT Introduction: Coronavirus pandemic began in China in 2019 (COVID-19), causing not only public health problems but also great psychological distress, especially for physicians involved in coping with the virus or those of the risk group in social isolation, and this represents a challenge for the psychological resilience in the world population. Studies showed that health professionals had psychological symptoms such as depression, anxiety, insomnia, stress, among others. Objectives: To investigate the quality of sleep and the prevalence rate of sleeping disorders among physicians during COVID-19 pandemic, and identify the psychological and social factors associated with the condition. Methods: A cross-sectional study of an online questionnaire was applied for physicians in Brazil. Among the 332 participants included, 227 were women. Sociodemographic assessment was used in the questionnaire, as well as the scale of impact on the events of modifications caused by COVID-19, assessment on sleep quality (PSQI), presence and severity of insomnia (ISI), depressive symptoms (PHQ-9), and anxiety (GAD-7). Results: Most physicians (65.6%) had changes in sleep. Poor sleep quality was reported by 73.1%, depressive symptoms were present in 75.8%, and anxiety in 73.4%. Conclusion: Our study found that more than 70% of the physicians assessed had impaired sleep quality, characterizing insomnia symptoms during COVID-19 outbreak. Related factors included an environment of isolation, concerns about COVID-19 outbreak and symptoms of anxiety and depression. Special interventions are needed to promote health professionals’ mental well-being and implement changes in this scenario.


INTRODUCTION
Coronavirus was identified in China in the late 2019's as a responsible agent of infection called coronavirus disease 2019 (COVID- 19), recognized as a worldwide pandemic by the World Health Organization (WHO) in March 2020, with a high potential of contagion and catastrophic incidence 1 . Up to September 4, 2020, the virus has infected more than 26 million people, causing more than 860,000 deaths worldwide 2 .
Epidemiological factors related to the virus, such as its incubation period, geographic reach, number of infected people and the real mortality rate have led to insecurity and fear in the whole population in the world. The situation aggravated due to insufficient control measures and the lack of effective therapeutic approaches, in addition to the precarious public health infrastructure 1,3 . These uncertainties have had consequences in several sectors in the routine of people's life, with direct implications on individuals' mental health especially among health professionals as they had to deal with the extremely adverse context of this new disease 3 .
This high psychological burden has already been reported by health professionals in Wuhan, where the virus began, and in other regions in China 4 .
In view of the critical situation, health calamity, the health professionals, especially those who were directly involved with patients with COVID-19 in the diagnosis, treatment and care, were exposed to a greater risk of developing psychological distress and other symptoms, such as impaired mental health 5 . As an example, a multinational and multicenter study found a significant association between adverse psychological results and physical symptoms in health professionals during the current pandemic. Of the 906 participating health professionals, all had moderate to severe levels of suffering: 5.3% of depression, 8.7% of anxiety, 2.2% of stress, and 3.8% of adverse psychological suffering 6 .
There are data showing that health professionals involved in coping with the outbreak of COVID-19 are more likely to have poor sleep quality compared to other occupational groups 7 . The experience of this whole set of difficulties, as mentioned above, as well as the alarming information that generate fear and insecurity, can trigger difficulties in starting and continuing to sleep, resulting in a sensation of non-restorative sleep, characteristic symptoms of insomnia. Insomnia has a direct and increased association in this context of psychological distress, impairing the quality of sleep, becoming a negative incremental factor on individuals' physical and psychological health who are experiencing the pandemic 8,9 . This study is justified by the need to understand the impact of this enormous change promoted by the pandemic on the physicians' sleep patterns who are involved and coping with COVID-19, allowing preventive or therapeutic measures to be taken on these professionals, particularly those exposed to the effects of unusual working conditions, restrictions and insecurity. It is also justified by the fact that after the pandemic many of the insomnia conditions may persist, reinforcing unhealthy habits in relation to sleep, perpetuating an adverse state to physical and mental health.

Study design
Cross-sectional study involving 332 physicians participating or not on the frontline of suspected or confirmed cases of COVID-19 in Brazil. An online form was used from the Google Forms platform; by means of a link shared through social media, WhatsApp, e-mail or other online systems, physicians working in different health services (primary care services, urgencies and emergencies, tertiary services, medical residency groups, public or private hospitals) were invited to perform the assessment regardless of specialties and Brazilian socio-demographic situation.

Measuring instruments
Through the online questionnaire, we collected sociodemographic data and administer scales to assess sleep quality, presence and severity of insomnia, as well as the symptoms of depression and anxiety. The sociodemographic characterization was carried out through a structured questionnaire gathering information about age group, sex, ethnicity, residential location, household members ( family, spouse, friends, alone, others), if they had children, if they were attending medical residency and year, area of activity, presence of comorbidities (with specification), if they were in contact with COVID-19 in their work sector, if they were removed, relocated or belonged to any risk group.
In this study, we used two instruments considered essential to assess sleep according to the American Academy of Sleep Medicine: • Insomnia Severity Index (ISI) 10 -it has been used to screen insomniacs and to evaluate the effectiveness of treatments used in clinical and population studies, as it has excellent psychometric properties and the ability to differentiate individuals with and without insomnia. Each item is rated on a scale of 0 to 4 points, and the total score ranges from 0 to 28. A total score ≥ 8 is compatible with diagnosis of insomnia, referring to mild ( To assess anxiety and depressive symptoms, we used Patient Health Questionnaire-9 (PHQ-9) and Generalized Anxiety Disorder (GAD-7): • PHQ-9 12 -consists of 9 items used to measure symptoms of anxiety, depression, and each of the 9 items scores on a scale of 0 to 3. The total score suggests different levels of depression: minimal/no depression (0-4), mild (5-9), moderate (10)(11)(12)(13)(14) or severe (15)(16)(17)(18)(19)(20)(21). • GAD scale 13 -it is a simple instrument containing 7 items that is based on the Diagnostic and Statistical Manual of Mental Disorders-fourth edition (DSM-IV) criteria and can easily track anxiety symptoms. The total score can be categorized into 4 groups of severity: minimal/no anxiety (0-4), mild (5-9), moderate (10)(11)(12)(13)(14) or severe (15)(16)(17)(18)(19)(20)(21).
We assessed the pandemic impact with the Impact of Event Scale with modifications for COVID-19 (IES-COVID19) 14 . It shows to be valid and presented reliable measures that can be used to investigate stressed symptoms related to trauma associated with a short-and long-term impact. The instructions have been adapted to apply specifically in relation to COVID-19. We also addressed in this evaluation the presence of symptoms related to the pandemic, such as fear of being contaminated and transmitting the disease; safety in the workplace; use or increased use of tobacco, alcohol, marijuana and stimulating substances (amphetamine and cocaine); search for pandemic-related news; or perceived impairment of social relationships.

Data processing and analysis
The data collected from the physicians using the Google Forms were exported to the Statistical Package for the Social Sciences (SPSS) in 13.0 version, constituting the database for statistical analysis in line with the objectives of this study.
In order to characterize the researched sample, we performed a descriptive analysis of the data. We calculated the absolute (N) and relative (%) frequencies with the categorical/qualitative variables of interest, as well as graphs; in the case of quantitative variables, descriptive measures were obtained, such as minimum, median, mean, standard deviation (SD) and maximum. We performed the chi-square test to verify a statistically significant association between sleep quality and the presence of depression and anxiety. We considered variables with p<5% as being significant.

Ethical considerations
Anonymity and confidentiality of data were ensured. The research project was carried out in accordance with the provision of the Resolution 466/12, reviewed and subordinated to the National Health Council for conducting research on Human Beings, and approved by the Research Ethics Committee. The agreement of the Informed Consent Form (ICF) was a mandatory item to continue the participation and contained the necessary information to clarify the research proposal, as well as voluntary participation, risks, benefits and data related to the Research Committee.
Of the participants, 60.2% were considered white, 32.8% were mixed and 5.4% were black. They had some comorbidity (47.4%), being the psychiatric -including depression, generalized anxiety disorder, panic disorder, bipolar disorder, among others -more prevalent (13.3%), and obesity (6.3%). The physicians were divided into 27 specialties: surgical specialties (10.5%), general practitioners (9.3%), family and community medicine (10.8%), emergency/intensive care (8.7%), Pediatrics (13.6%), Neurology (14.5%) and other clinical specialties (5.1%). Of the responses obtained, 28% were physicians who were attending the medical residency program. Approximately 88% of the physicians worked in a sector with the possibility of contacting patients with COVID-19. Of these, 31% had their function modified to meet the demand to care for patients with COVID-19 and 3% were removed from their duties because they belonged to a risk group.
On the IES-COVID19, 64.1% had fear of having or transmitting the disease, and 61% felt unsafe at the work environment. Regarding to the consumption of alcohol and tobacco, 36.2% showed an increase in the consumption of alcoholic beverages, and 4% started to use or increased their tobacco habit. There was a 4.2% of increase in the consumption of stimulants, such as cocaine or amphetamines.
According to the data in Table 2, the study revealed that most physicians (65.6%) had complaints in relation to sleep, with symptoms compatible with mild (37.2%), moderate (23.3%) and severe insomnia (5.1%), according to the ISI. Among the main aspects questioned, 74% reported difficulty in initiating sleep, 66.5% had problems with early awakening, and 66.2% difficulty in continuing to sleep (Figure 1).  In addition, 61% had some degree of dissatisfaction with the current sleep pattern, 49.2% believed that their problem with sleep was noticeable to other people, and 75.8% believed that their problem of sleep interfered with their daytime activities due to daytime fatigue, some degree of impairment in their ability to work/perform daily activities, concentration, memory, mood (Figures 2, 3 and 4, respectively). The physicians' average sleep time score was 6 hours within 24 hours (±1.3 SD), with an average latency time to sleep onset in 46.8 minutes (±47.8 SD) and 7.5 hours (±1.6 SD) of staying in bed at night (not necessarily sleeping).
In the subgroup of resident physicians (28%), 65.6% had symptoms of insomnia according to the ISI -exactly the same percentage obtained from the analysis of the total sample of participants, and 26.9% reached scores classified as moderate to severe insomnia (Table 3).

DISCUSSION
It is known that the main contributors to the emergence or potential worsening of a situation of poor sleep quality are the increased levels of stress and sudden changes in the sleep routine 15 . Several factors are possibly associated with physical and mental stress to which health professionals are exposed in pandemic situations such as COVID-19: the medical team often needs to have personal protective equipment (PPE) for 12 hours or more on duty, including doublelayer protective equipment, masks, gloves, cloaks, caps and goggles 16 . In addition to the feeling of insecurity at the work environment, as well as increased consumption of stimulating substances, alcohol and/or tobacco were pointed out in our study as contributing factors to the symptoms of anxiety, depression and poor sleep quality, with consequent repercussions on insomnia severity.
Our data, as well as those in the literature, show that these professionals usually experience stress in the context of pandemics, namely: increased risk of being infected, getting ill and dying; possibility of inadvertently infecting others; overload and fatigue; frustration for not being able to save lives, despite efforts; threats and aggression, perpetrated by people who seek care and cannot be accepted due to limited resources; and separation from family and friends 17 . Regarding to COVID-19 in particular, the challenges faced by physicians can trigger or intensify symptoms of anxiety, depression and stress, especially when it comes to those working on the so-called "front line" 18 . In this study, a significant proportion of participants experienced the symptoms of anxiety (75.8%), depression (73.1%) and poor sleep quality (73.4%). In general, these professionals have been discouraged to interact closely with other people, which tends to increase the feeling of isolation; and, still, they usually dedicate significant time of their day to put on and remove personal protective equipment, which increases the exhaustion related to work 19,20 . Faced with these conditions, the medical team is mentally and physically exhausted, and therefore has an increased risk of insomnia due to the high level of stress during the day.
In addition, sleeping disorders can have direct consequences on the emotional functioning on the next day. Work overload and symptoms related to stress make health professionals very vulnerable to psychological distress, which increases the chance of developing psychiatric disorders and insomnia symptoms 21 , as it had occurred with some of the health professionals who participated in this study during the pandemic.
We found that almost half (47.4%) of the physicians reported some previous comorbidity. Among those mentioned, psychiatric (13.3%) were the most mentioned, highlighting depression, generalized anxiety disorder, panic disorder and bipolar disorder. We consider these symptoms relevant, as psychiatric conditions are known to be an important contributing factor for sleep impairment and insomnia in relation to healthy ones 21 ; and more than 25% of the psychiatric patients have the same symptoms as those of post-traumatic stress and moderate to severe insomnia 22 .
Interventions with an emphasis on mental health, including the work of psychologists and psychiatrists can contribute to the strengthening of the support network, by encouraging them to maintain frequent contacts, during breaks at work, through phone calls, text messages, audio and video 23,24 . This also tends to benefit the mental health of people in the support network of health professionals, as keeping them informed may reduce negative emotions, such as fear and anguish 24 . Insomnia is the main complaint of sleep in terms of incidence and recurrence in the day-to-day of the most varied health professionals and has a major negative impact on the quality of life of these individuals, especially when it becomes chronic, increasing morbidity and mortality, performance in personal, work and social life, generating high direct and indirect costs 25,26 . Despite its great clinical relevance, insomnia is not diagnosed and evaluated in a systematic way, a fact well illustrated by the great variability in its prevalence reported in the literature 27,28 . Insomnia and chronic sleep deprivation represent an independent risk factor for psychiatric disorders such as ideation, attempts and suicidal death 25 . Therefore, in our assessment, physicians not only had particularly high scores on the symptoms of depression, anxiety and insomnia, but we also found a repercussion of these symptoms on poor sleep quality and mental health, which requires special care.
It is essential that the psychological support and intervention measures involved in the context of a crisis also aim to offer in coping with strategies to deal with intrusive thoughts and anticipatory or situational anxiety 29 . It is important to highlight the complexity of the Brazilian health system, in which the health professionals who work in different levels of complexity may need different strategies on information, support or interventions. Moreover, historically, few mental health programs are available for health professionals 30 . There are few studies in Brazil that address epidemiological data and intervention models focused on health professionals' mental health who were involved in assisting patients with COVID-19. Understanding the problems encountered in the  insomnia group can help Health Service managers with effective education and training for mental health care. The present study aimed to draw attention to the relevance of the topic and suggests the implementation of both research and health care, especially for those with different psychosocial risk factors.
The possible limitations of this study are that some medical specialties were not included, and the scope to include frontline doctors or not was limited to the scope of sharing the questionnaire through social network, but participation in the study was completely voluntary, with an acceptable response rate. Another limitation of the Brazilian study was that it did not reach the 26 States of the federation.
This study revealed that most physicians experienced sleep-related problems during the COVID-19 outbreak. Sleep duration was considered insufficient and the quality unsatisfactory, in addition to the decreased physical and mental functioning during the day. We observed high rates on the symptoms of insomnia, anxiety and depression, contributing significantly to poor sleep quality. Adequate treatment for sleep disorders is always vital, contributing to physical and mental health, reducing vulnerability to worsen or onset psychiatric disorders and risk of suicide. Special interventions in promoting health professionals' mental wellbeing need to be implemented to change this scenario.