Abstract
BACKGROUND:
Atypical working hours have raised serious concerns about health effects, such as sleep disorders and psychological repercussions. These schedules are frequent among health professionals having the obligation to ensure the permanence of health services.
OBJECTIVE:
To assess the impact of atypical working hours on sleep, sleepiness, and mood among health personnel (HP).
METHODS:
Comparative descriptive cross-sectional study was carried out among HP of a regional hospital in Southern Tunisia from December 2019 to May 2020. Two groups were defined according to their work schedule: group A had a fixed-day work and group B had atypical working hours. The assessment of sleep, sleepiness, and mood was based respectively on the Spiegel questionnaire, the Epworth Sleepiness Scale, and the Hospital Anxiety and Disorder scale.
RESULTS:
A total of 101 HP in group A and 135 in group B were included. We found signs of pathological sleep in group B: greater delay in falling asleep (78.1% Vs 53.5%; p = 0.002), altered sleep quality (88.1% Vs 56.4%; p = 0.039), and more frequent occurrence of dreams (37% Vs 31.7%; p = 0.033). The quality of sleep was associated with the professional seniority (p = 0.01), the workplace (p = 0.02), and having dependent children (p = 0.04). The mean score of depression was higher in group B (p = 0.02). Depression was associated with the workplace (p = 0.04) and the quality of sleep (p = 0.01).
CONCLUSION:
The results of this study indicate that atypical working hours generate both sleep and mood disorders in HP. It is vital to encourage the occupational physicians to estimate sleep and mood disorders in HP.
Introduction
Atypical working hours are defined by the French National Research and Safety Institute for the Prevention of Occupational Accidents and Diseases [1] as “all working time arrangements that are not “standard”, knowing that standard working corresponds to work configurations composed of “5 regular days per week from Monday to Friday, with working hours between 5 and 11 hours per day, and 2 days off per week. The most-known atypical working hours are shift work, night work, and weekend work [1, 2]. In the definition by the National Institute of General Medical Sciences, the circadian rhythms are physical, mental, and behavioral changes that follow a 24-hour cycle [3]. For the purpose of this article, atypical working hours are preferred on circadian rhythm since the participants do not necessarily follow the 24-hour cycle in their work schedules. The results of a previous review indicated that abnormal patterns of sleeping could lead to immunological issues, hyper-tension, metabolic syndrome, insomnia, cardiovascular disease, obesity and depression [4]. In addition, for several years, atypical working hours have raised serious concerns about various health effects, such as sleep and sleepiness disorders as well as psychological repercussions [2, 5]. These atypical schedules are frequent among health professionals (HP) having the obligation to ensure the continuity of work, the safety of people, and the permanence of health services [6]. In this context, we conducted a survey among HP in a hospital in southern Tunisia, to assess the impact of working atypical working hours on sleep, sleepiness, and mood in this staff, which is to our knowledge the first study in Tunisia to focus on this topic among these first-line fighters of health systems.
Methods
Study design
This was an observational cross-sectional and descriptive study.
Setting
The study was carried out from December 2019 to May 2020 among HP of a Regional Hospital in southern Tunisia. Data collection was based on a self-administered questionnaire including socio-demographic, professional, and medical characteristics.
Participants
The participants of this study were HP with at least one year of professional seniority. HP hired for less than one year and those who were on sick leave during the study were not included. HP who refused to answer the questionnaire were excluded. Two groups of participants were defined according to their work schedule: the group A with fixed-day work and the group B with atypical working hours (fixed night work and 2*8 or 3*8 shift work). To assess sleep and mood disorders, we used the definition established by the INRS (French National Research and Safety Institute for the Prevention of Occupational Accidents and Diseases) [1] to group fixed night work and shift work as atypical working hours.
Variables
The following data were extracted: socio-demographic characteristics (age, gender, marital status, and number of dependent children or parents), professional characteristics (professional seniority, occupational grade, and work department) and medical history. The quality of sleep and the level of daytime sleepiness were measured using valid scales. Mood disorders were assessed in the two groups.
Data sources/measurement
The quality of sleep was assessed by the Spiegel questionnaire which included six questions related to sleep onset time, sleep quality, sleep duration, nocturnal awakenings, and dreams to obtain a total score of 30 points [7–9]. The quality of sleep is classified as normal when between 24 and 30 points, disturbed when between 15 and 23 points, and pathological below 15 points. We used the Epworth Sleepiness scale [10] to assess three levels of daytime sleepiness according to the score obtained: absence of sleep debt (score≤8), presence of sleep debt (score between 9 and 14), and presence of signs of excessive daytime sleepiness (score≥15). Mood disorders were measured using the HAD scale (Hospital Anxiety and Depression scale) [11], which allowed the calculation of both the depression and the anxiety scores with thresholds determining three groups: A score less than or equal to 7 stands for no anxiety or depressive state, a score between 8 and 10 indicates doubtful anxiety or depressive state, a score greater than or equal to 11 means certain anxiety or depressive state.
Bias
The self-administration of the questionnaire could also generate reporting bias. Incomplete answers to the questionnaire were excluded.
Study size
A total of 236 complete responses to the questionnaire were collected.
Statistical methods
Descriptive statistics were used to summarize the entire cohort and to compare HP in the two groups. The characteristics of the participants are described as proportions for categorical variables and means and standard deviations for continuous variables with a normal distribution. Differences between groups was assessed was assessed by the Chi square test and -Fisher’s exact test for categorical variables. The Student t-test was used for analysis of continuous variables with normal distribution and the Mann–Whitney U test for analysis of continuous variables with skewed distribution. The significance level was set at 0.05. A multivariate logistic regression model was not established. The data were analyzed using Software Package for Statistics and Simulation (IBM SPSS version 25, IBM Corp, Armonk, NY, USA).
Ethical considerations
At the time of our study, the approval of the Ethics Committee or the institutional review board was unfortunately not provided. This study, as an observation of the outcomes of work schedules on HP, is exempt from Institutional Review Board approval [12] and met the following criteria: The information obtained is recorded by the investigator in such a manner that the identity of the Human Subjects cannot be readily ascertained, directly or indirectly through identifiers linked to the subjects.
Results
Descriptive participants’ characteristics
A total of 236 HP responded to the questionnaire, with a response rate of 69.4% (236/340). The mean age was 38.95±9.26 years. Our population was composed of 130 women and 106 men (sex ratio of 0.8). They were married in 70.8% and had dependent children and dependent parents respectively in 64% and 43.6% of cases. They had a mean professional seniority of 12.72±9.3 years. The most represented occupational grade was that of nurses (38.6%). The participants worked mainly in the emergency department, medical departments and surgery departments in respectively 22.9%, 18.6%, and 15.3% of cases. Our respondents declared that they were smokers in 20.8% of cases (n = 49), and practiced regular physical activity in 21.2% (n = 50). Pathological medical history was noticed in 31.4% of cases (n = 74) represented mainly by hypertension (n = 19), diabetes (n = 13), and psychiatric illness (n = 7).
Comparison of socio-professional characteristics according to the work schedule
According to the work schedule, group A with fixed-day work was composed of 101 HP (42.8%) and group B (n = 135; 57.2%) with atypical working hours was composed of fixed night work (n = 58; 24.5%) and 2*8 or 3*8 shift work (n = 77; 32.6% of all cases).
The socio-professional characteristics of the HP according to their work schedule are shown in Table 1. No significant difference was observed in age between the two groups of participants (p = 0.08). Compared by gender, women had more fixed-day work (63.3% Vs 48.8%, p < 0.05). The proportions of HP with a pathological medical history, dependent children or parents did not differ between the two groups. The distribution of the participants according to the marital status, and the professional seniority did not show any significant difference between the two groups. The workers from the second group were likely more tobacco consumers (22.2% Vs 18.8%, p = 0.05). The practice of physical activity was higher among workers with atypical work schedules with no significant difference (p > 0.05). The work schedule was associated to the workplace and the occupational grade (p < 0.05). Atypical working hours were essentially seen in the emergency departments (33.3%) and medical departments (20.7%). The participants were mainly nurses and health technicians in group B (46.7% and 21.5%) as shown in Table 1.
Socio-professional characteristics of the participants according to the work schedule
Socio-professional characteristics of the participants according to the work schedule
n: Number; % : Percentage; *: likelihood ratio.
Sleep quality
The mean Spiegel score was 18.8±4.4 with a significant difference between the two groups (p = 10–3). By identifying the three sub-scales of the Spiegel score, we found that HP with atypical working hours showed signs of pathological sleep (23.7% vs. 8.9%, p = 0.01) compared to the fixed-day workers (Table 2).
Work schedule effects on sleep, sleepiness and mood in the two groups
Work schedule effects on sleep, sleepiness and mood in the two groups
n: Number; % : Percentage.
We found a greater delay in falling asleep in 78.1% of cases from group B compared to 53.5% from group A (p = 0.002), altered sleep quality (88.1% Vs 56.4%; p = 0.039), and more frequent occurrence of dreams (37% Vs 31.7%; p = 0.033) (Table 3). In addition, we have found an association between the quality of sleep and the following factors: the professional seniority (p = 0.01), the workplace (p = 0.02), and having dependent children (p = 0.04).
Sleep quality characteristics in the two groups according to the Spiegel scale
n: Number; % : Percentage.
The mean Epworth score was 8.14±4.4 with no significant difference between the two groups of workers. By identifying the three sub-scales of the Epworth score, it was found that both groups showed almost identical signs of excessive daytime sleepiness (9.6% vs. 9.6%, p = 0.8). Moreover, sleep debt was paradoxically observed in the first group as shown in Table 2.
Mood disorders
The mean score of the HAD scale relating to anxiety was 8.9±3.88 with no significant difference between the two groups. Anxiety was severe and almost in identical proportions among the two groups (31.7% in group A Vs 31.1% in group B, p = 0.5). A significant association was found between anxiety and occupational grade (p = 0.04), and severe anxiety was observed mainly in nurses (56% of all severe anxiety cases).
The mean score of the depression scale was 7.6±3.39. Depression was severe according to this score and higher in group B (31.9% Vs 22.8%, p = 0.02) (Table 2). A significant association between depression and the workplace (p = 0.04) was also found. We also highlighted a significant relationship between depression and sleep quality (p = 0.01) as shown in Table 4.
Relationship between depression and sleep quality (N = 236)
Relationship between depression and sleep quality (N = 236)
n: Number; % : Percentage.
To our knowledge, this study is likely the first to focus on the impact of atypical working hours on sleep and mood disorders among healthcare workers of a second-line healthcare facility in Tunisia. Sleep quality was assessed through the Spiegel questionnaire which makes it possible to estimate the quality and quantity of sleep in health personnel suffering from sleep disorders and therefore to help in early diagnosis. Its intrinsic validity makes it a good test due to the reliable reproducibility of measurements [9]. The Epworth Sleepiness Scale, which is the most used simple and precise subjective tool, allowed us to measure the general level of daytime sleepiness [10]. Mood disorders were assessed through the HAD (Hospital Anxiety Depression) scale, which validity is confirmed by several studies when in community and healthcare settings to detect depression and anxiety [11]. This study has some limitations, particularly when defining the two groups. Indeed, the second group with atypical schedules seemed to be heterogeneous because of its composition involving certain parameters such as the fixed night work and the possibility of day work in the shift work. This heterogeneity could be seen in the results obtained. The cross-sectional nature of the study also prevented a long-term follow-up which is preferred by researchers in occupational health as Golberg et al. in their study due to the temporality between exposure and disease onset. In fact, these analytical studies help more to assess the role of risk factors on health and to limit inclusion bias due to the voluntary nature of the study [13]. The self-administration of the questionnaire could also generate reporting and memorization bias. Also, a pre-test was not carried out to detect response bias since it was the first study conducted in this context in this type of healthcare facility [14].
According to our results, the predominance of sleep disorders among HP with atypical working hours suggest that this work schedule could generate more sleep disorders than day work. In a review of the literature, the French National Institute for Research and Safety [15] showed the effects of shift and/or night work on sleep, particularly a decrease in total sleep time per 24 hours leading to a chronic sleep debt and an increase in the risk of sleepiness during the waking period.
Atypical work hours are not physiological but the adaptation of workers to this type of work schedule is linked to the balance between three factors: a chronobiological factor, a sleep factor and a domestic factor. The first factor which is the chronobiological factor is explained by Leger et.al in their review about medical consequences of shift work on sleep. They claimed that the atypical schedule and especially shift work causes a desynchronization of the circadian rhythm, thus disturbing vigilance, sleep, and quality of life [16]. This is on the line with the results of our study; we found that sleep was disturbed and pathological in staff working atypical hours. The second factor is sleep factor and corresponds here to an altered duration and quality of sleep. One of the most documented effects of atypical schedules on health is sleep debt as shown by Adam et al. in their study including 1470 German and 1231 French workers divided into four groups according to their work schedules, they found that the day shift group complained of sleepiness after the morning shift and significantly more than the day or night workers [17]. This finding is in agreement with Ohayon’s (2010) and Zhou’s (2017) findings which showed that working early in the morning or late at night cuts off the main sleep period and therefore reduces not only the duration but also the quality of sleep [18, 19]. In accordance with these results, previous studies have demonstrated that chronic sleep debt with insufficient sleep for several days or weeks leads also to drowsiness and sleepiness [16, 20–22]. Concerning the quality of sleep, the results of this study are in keeping with previous observational studies, in which several sleep disturbances have been observed among shift workers, such as difficulty in falling asleep, frequent sleep and early morning arousal [15, 23]. To remedy this, we most often resort to daytime sleep, which is less restorative than nighttime sleep. It is often shorter and disrupted by unfavorable environmental conditions (noise, light, temperature) [24–27]. The third factor, the domestic factor, is known as a tolerance to shift work provided that a favorable home environment exists. Environmental conditions maladaptive to sleep are mainly daylight during rest, higher temperature and noise levels during the day, and social obligations. Social and family disruptions are often present because of atypical schedules [16]. In this study, having dependent children had a significant relationship with sleep disorders.
Sleepiness disorders in employees with atypical working hours are related to two factors in previous studies [2, 28]. The main factor is the desynchronization of the biological clock leading to poor adaptation to the circadian rhythm changes, difficulty in falling asleep and increased drowsiness during waking hours [23, 29–30]. The second factor is that the sleep debt caused by atypical working hours leads to daytime sleepiness as a physiological recovery [29, 31]. Sleepiness is also recognized as one of the cardinal symptoms of “shift work intolerance syndrome” (SWSD: Shift Work Sleep Disorder), according to the international classification of sleep disorders [32].
Contrary to expectations, in this study we found that day workers had a slightly greater sleep debt compared to the workers with atypical schedules and this could be explained by the amputation of the main sleep period by early morning work [15]. Indeed, early awakenings and reduced sleep time are correlated with increased sleepiness and frequent napping resulting in a decrease in cognitive and physical performances during the rest of the day and even more during the night according to the report of the collective expertise of the French Agency for Food, Environmental and Occupational Health & Safety (ANSES) in 2016 [5]. However, the observed association between sleepiness and the work schedule was not significant in both groups of our study. This result could be explained by the heterogeneity of the second group composed of HP occupying a chosen fixed night shift to which they could easily adapt, and those who have shift work that offers the possibility of recuperation during the other periods of the day.
In addition to sleep disorders, mood disorders such as depression and anxiety appear to be directly [33, 34] or indirectly [35, 36] linked to atypical working hours. Our study revealed a significant relationship between depression and work schedule which is similar to the findings of previous studies in Korean and Japanese workers by Lee (2016), Park (2016) and Togo (2017) respectively [37–39]. Contrary to expectations, our study did not find a significant difference between anxiety and work schedule, but at least a significant association was noted between anxiety and occupational grade represented mainly by nurses having certain anxiety. We have also demonstrated a significant association between depression and the workplace, similar to several studies carried out among doctors [40, 41] and nurses or caregivers [42] in health departments, particularly in emergencies [40]. Several explanations for the relationship between depressive symptoms and work-related risk factors among nurses have been cited by Gong and colleagues in their cross-sectional study, such as the difficulty of clinical activity in specific departments, the stressful working conditions with staff shortages and workplace violence [43]. Also, we highlighted a significant relationship between depression and sleep, which is in the line with the results of previous studies [44–46].
We can consider that atypical working hours could influence sleep and mood, leading to sleep debt, alteration in the quality of sleep, depression, and anxiety. Preventive measures are needed to face these deleterious effects, such as an arrangement of the work schedules by proposing a voluntary fixed night or day work, schedule with a fixed number of working hours that can not be exceeded unless necessary, and providing compensatory rest.
Conclusion
The results of this study indicate that atypical working hours generate both sleep and mood disorders in health personnel. It is vital to encourage the occupational physicians to estimate the quality and quantity of sleep and mood disorders in these workers and thus assist in early diagnosis and management. These results encourage us to take into account all the physiological, psychological and organizational factors to seek preventive strategies to improve work conditions and quality of life of the health personnel.
Ethical considerations
Not applicable.
Informed consent
Participants were informed of the objectives of the study and their consent was obtained. The anonymity of the respondents was respected.
Conflict of interest
The authors declare that they have no conflict of interest.
Footnotes
Acknowledgments
The authors have no acknowledgments.
Funding
The authors report no funding.