Keywords
occupational risks - sleep - occupational health - working conditions - healthcare
workers - pyschosocial factors
Introduction
The increasing globalization and financialization of the economy in recent years have
had a profound effect on the occupational environment and the lives of workers. Precarious
working conditions, such as the violation of labor rights, the lack of safety in workplaces,
and the increase in the work pace, have intensified the demands and pressure placed
on workers.[1] This has become even more pronounced in the context of the coronavirus disease 2019
(COVID-19) pandemic, with an increase in the flexibilization of contracts and relationships
in a phenomenon denominated the “uberization” of the healthcare workforce (e.g., temporary
contracts, outsourcing contracts through staffing agency, etc.). This phenomenon deteriorates
working conditions, with the new home-office modality, an increased workload, and
deregulated workers' health protection, as demonstrated by the scarcity of materials
and protective equipment.[2]
[3]
The COVID-19 crisis intensified preexisting psychosocial factors in healthcare work.[4] Such factors can cause psychological or physical harm to workers[5] and are related to the work environment and organization, interpersonal relationships,
worker health, and exposure to offensive behaviors, such as sexual harassment and
workplace violence.[6] This context is also marked by the lack of personal protective equipment (PPE),
the redeployment of professionals for the treatment of infected patients, the moral
harm arising from the management of scarce resources and setbacks for decision making,
failures in communication, the increased workload, the increase in violence, and financial
insecurity.[4] Thus, the prevalence of anxiety, stress, and depression among healthcare workers
increased, especially among those who worked on the frontlines of the COVID-19 pandemic,[4] along with an increase in absenteeism due to mental illness.[7]
Besides precarious working conditions, another factor aggravated by the pandemic was
the quality of the worker's sleep, which is a reflection of the increase in the workload
to cope with the absenteeism of staff members and the high demand of patients as well
as the situation of insecurity and crisis.[2] As services in the health field operate uninterruptedly, night shifts are commonplace.
However, nightshift workers experience the interruption of the normal sleep pattern
due to the asynchrony between the workers' activities and changes in sunlight, compromising
the circadian rhythm. The change in the sleep pattern can exert a negative impact
on the quality of sleep and workers' health—both psychologically and cognitively,
also affecting one's work performance and the quality of the care provided.[8]
[9]
Healthcare professionals who work in the night shift have lower quality sleep compared
to dayshift workers[10]
[11]; those on shift work had worse quality of sleep and developed more sleep disturbances
than non-healthcare professionals during the COVID-19 pandemic.[12] The shift work is related to adverse physiological and immunological consequences
for health, such as vitamin D deficiency,[13] which can favor COVID-19 infection even more so in workplaces with favorable transmissibility
conditions by the high flow of people.[14]
Short sleep duration is associated with metabolic syndrome, hypertension, obesity,
sleep disorders, fasting glucose and immunological changes,[15] which are considered risk factors for worsening COVID-19 infection outcomes such
as hospitalization, invasive mechanical ventilation, and even death.[14] Moreover, insomnia and short sleep duration are prevalent and are associated with
psychological distress, with a higher prevalence of symptoms of acute stress, depression,
and anxiety in this population.[16]
Studies have found a correlation between psychological distress and quality of sleep
in the pandemic context[16]
[17]
[18]
[19]
[20] and identified increased prevalence of sleep problems, anxiety, burnout, and depression,
as well as the risk factors and the predictors of poor sleep quality and mental health
diseases across this population.[17]
[18]
[20] Associations between psychological distress and short sleep,[16] sleep probems,[19] and sleep quality levels[20] have also been also found.
Several studies have evaluated psychosocial factors in the work environment and the
quality of sleep among healthcare workers, including the association between psychological
distress and sleep quality. However, no studies were found that correlated workplace
psychosocial factors (which increase the risk of occupational stress[21]) and quality of sleep, especially considering the particularities of each healthcare
professional category and the context of the pandemic.
Therefore, the aim of the present study was to investigate psychosocial factors at
work, sleep characteristics, and the correlation between these aspects in healthcare
workers. Such results can highlight the psychosocial factors that have a greater impact
on sleep quality and assist in the development of effective strategies for improving
the quality of sleep as well as preventing illness and the degradation of work-related
quality of life.
Materials and Methods
Study Design
A cross-sectional study (e-survey) was conducted following the Checklist for Reporting
Results of Internet E-Surveys (CHERRIES).[22] This study integrates the longitudinal research HEalth conditions of healthcaRe
wOrkErS (HEROES),[23] the aim of which was to investigate psychosocial aspects in the workplace, sleep
characteristics, musculoskeletal symptoms, and depression among healthcare workers
of the Brazilian healthcare system, a universal free access system to all Brazilian
citizens.[24]
Sample
The sample consists of 125 healthcare workers from the HEROES cohort.[23] The inclusion criteria were being a public healthcare worker, aged 18 years or older,
and working in healthcare activities during the COVID-19 pandemic. Participation in
the study was voluntary, and no financial incentive was offered. Students, retirees,
duplicate responses, and inconsistent data were excluded.
Participants recruitment was carried out on internet channels as well as through the
press, social networks, and e-mails available on institutional websites. The researchers
publicized the project through interviews on local radio stations, articles in the
press as well as profiles on Instagram, Facebook, and YouTube. Emails were also sent
to public hospitals, Secretaries of Health, as well as health services and councils.
One hundred forty-three workers answered the questionnaire, 125 of whom met the inclusion
criteria and comprised the convenience sample. The reasons for exclusion were not
working in the health field at the time (n = 10), duplicate answers (n = 4), and not
being a worker in the public healthcare system (n = 4).
The sample size was calculated a posteriori, using the G*Power software.[25] The point biserial correlation test was chosen, and the calculation considered a
type I error of 5%, a power of 80%, and an effect size of 0.18. The required sample
was 237, but we only reached 53% of participation, which is a limitation of our study.
Data Collection
Data gathering was done via Google Forms (Google LLC., Mountain View, CA, USA) to
comply with the contact restrictions imposed by the COVID-19 pandemic in the period
from June 19, 2021, to April 4, 2022. Three instruments were used for data collection:
(i) a sociodemographic and occupational questionnaire containing questions related
to gender, age, marital status, education, health conditions, lifestyle habits, and
occupational history; (ii) the short version of the Copenhagen Psychosocial Questionnaire
II[6]
[26] validated for Brazilian Portuguese (COPSOQ II-Br), with Cronbach alpha values between
0.70 and 0.87[6]; and (iii) the Pittsburgh Sleep Quality Index (PSQI)[27]
[28] validated for Brazil (PSQI-Br), with Cronbach alpha of 0.82.[27]
Pretests were performed to determine the usability and technical functioning of the
questionnaires, estimate the response time, and correct typographical errors. The
statement of informed consent was included among the forms, and a copy signed by the
project coordinator was available for download.
Psychosocial conditions in the work environment were investigated using the short
version of the COPSOQ II-Br, which is composed of 40 items divided among 7 domains:
1. Demands at work; 2. Work organization and job contents; 3. Interpersonal relationships
and leadership; 4. Work-individual interface; 5. Workplace values; 6. Health and wellbeing;
and 7. Offensive behaviors.[6] The questionnaire is scored using a Likert scale, and the score is calculated in
accordance with the number of questions in each dimension (0–3 points, 0–4 points,
0–6 points, and 0–8 points). The scores enable the following classification: favorable health situation (green), intermediate situation (yellow), and health risk (red).[6] In the present study, the scores were dichotomized as no risk (favorable health status) and at risk (intermediate situation and health risk), what we called risk zone.
The PSQI-Br[29] is used to investigate sleep quality in the previous month by combining quantitative
and qualitative information on sleep and classifies respondents as good or poor sleepers. This questionnaire is composed of 19 self-administered questions grouped
into 7 components with weights distributed on a scale from 0 to 3: (i) subjective
sleep quality, (ii) sleep latency, (iii) sleep duration, (iv) habitual sleep efficiency,
(v) sleep disturbances, (vi) use of sleep medications, and (vii) daytime dysfunction.[27]
[29] The scores are summed to produce a total ranging from 0 to 21, with higher scores
denoting poorer sleep quality. An overall score greater than five points indicates
that the individual has difficulties in at least two components or moderate difficulties
in more than three components.[27]
[29]
Statistical Analysis
Only fully completed questionnaires were analyzed. TheIBM SPSS Statistics for Windows,
version 26.0 (IBM Corp., Armonk, NY, USA) was used for the descriptive analysis of
the variables of the three questionnaires, with the calculation of absolute (n) and
relative (%) frequencies and mean and standard deviation (SD) values.
Correlations between psychosocial aspects and sleep quality were investigated using
the biserial correlation test (rpb) since sleep quality was analyzed based on the total score (quantitative discrete
variable varying from 0–21). Psychosocial factors were analyzed considering the with risk (value 1) and without risk (value 0) zones (dichotomous variable). The significance level was set at 5%. Correlation
coefficients were interpreted as strong (rpb > 0.50), moderate (rpb between 0.30 and 0.50), or weak (rpb < 0.30)[30].
Ethical Concerns
This study met the ethical requirements for research involving human beings stipulated
in Resolutions 466/2012 and 510/2016 of the National Board of Health and received
approval from the Human Research Ethics Committee (decision number: 39705320.9.0000.5504).
All participants provided informed consent before completing the questionnaires.
Results
This study included 125 healthcare workers from the following regions of Brazil: Southeast
(79.2%), South (11.2%), Northeast (4.8%), Midwest (3.2%), and North (1.6%). Most were
female, with a mean age of 37.5 years, self-declared white, married, with a graduate
level of education, and without children ([Table 1]).
Table 1
Sociodemographic characteristics of healthcare workers (n = 125). Brazil, 2021 to
2022.
|
Characteristics
|
n (%)
|
|
Average age (years)*
|
37.5 (8.3)
|
|
Sex
|
|
|
Female
|
104 (83.2)
|
|
Male
|
21 (16.8)
|
|
Color/race
|
|
|
White
|
89 (71.2)
|
|
Brown
|
29 (23.2)
|
|
Asian
|
1 (0.8)
|
|
Black
|
6 (4.8)
|
|
Marital status
|
|
|
Single
|
41 (32.8)
|
|
Married
|
71 (56.8)
|
|
Widower
|
2 (1.6)
|
|
Separated/divorced
|
11 (8.8)
|
|
Education
|
|
|
Primary school
|
2 (1.6)
|
|
High school
|
22 (17.6)
|
|
Higher education
|
22 (17.6)
|
|
Graduate school
|
79 (63.2)
|
|
Number of children
|
|
|
None
|
65 (52.0)
|
|
One
|
30 (24.0)
|
|
Two or more
|
30 (24.0)
|
*Data expressed as mean (SD).
The participants were mostly nurses, working in hospitals, with a working time between
2 and 5 years, a 40-hour work week, employment contract governed by the Consolidation
of Labor Laws, income from 3 to 6 times the monthly minimum wage, and no more than
one employment contract ([Table 2]).
Table 2
Occupational characteristics of healthcare workers (n = 125). Brazil, 2021 to 2022.
|
Characteristics
|
n (%)
|
|
Occupation
|
|
|
Dentist
|
4 (3.2)
|
|
Physiotherapist
|
26 (20.8)
|
|
Nurse
|
45 (36.0)
|
|
Nursing technician/assistant
|
28 (22.4)
|
|
Physician
|
8 (6.4)
|
|
Other
|
14 (3.2)
|
|
Workplace
|
|
|
Primary care
|
40 (32.0)
|
|
Hospital
|
61 (48.8)
|
|
Emergency care
|
12 (9.6)
|
|
Outpatient care
|
4 (3.2)
|
|
Psychosocial care
|
5 (4.0)
|
|
Home care
|
3 (2.4)
|
|
Working time
|
|
|
Less than 6 months
|
7 (5.6)
|
|
Between 6 and 12 months
|
30 (24.0)
|
|
Between 1 and 5 years
|
42 (33.6)
|
|
Between 6 and 10 years
|
22 (17.6)
|
|
More than 10 years
|
24 (19.2)
|
|
Weekly workload
|
|
|
Less than 30 h
|
6 (4.8)
|
|
30 h
|
30 (24.0)
|
|
36 h
|
21 (16.8)
|
|
> 40 h
|
68 (54.4)
|
|
Type of contract
|
|
|
CLL*
|
60 (48.0)
|
|
Civil servant
|
52 (41.6)
|
|
Service provider
|
11 (8.8)
|
|
Outsourced
|
2 (1.6)
|
|
Income
|
|
|
More than 1 to 3 x MMW†
|
25 (20.0)
|
|
More than 3 to 6 x MMW†
|
49 (39.2)
|
|
More than 6 to 9 x MMW†
|
22 (17.6)
|
|
More than 9 x MMW†
|
25 (20.0)
|
|
Preferred not to say
|
4 (3.2)
|
|
More than one employment relationship
|
39 (31.2)
|
*CLL = Consolidation of Labor Laws; †MMW = monthly minimum wage in 2020 (US$ 220.64).
Most participants reported good sleep quality, except for nursing technicians/assistants,
who reported poor sleep quality. Sleep latency (time between lying down in bed and
falling asleep) was between 16 and 30 minutes among most participants, except for
nurses and nursing technicians/assistants, who reported sleep latency between 31 and
60 minutes and more than 60 minutes, respectively. Sleep duration varied among the
participants and was less than 5 hours among physicians (38%) and nursing technicians/assistants
(29%). Sleep efficiency was adequate among all categories of workers.
Sleep disturbances were reported less than once a week or 1 to 2 times a week by most
participants, with nurses and nursing technicians/assistants reporting disturbances
1 to 2 times a week. The use of sleeping medications was reported mainly by physiotherapists,
nursing technicians, and nurses. Most participants reported daytime dysfunction (difficulty
staying awake during daily activities, such as driving, eating, or participating in
social activities; and problems maintaining one's mood during usual activities) less
than once a week and/or mild, except nursing technicians/assistants and workers classified
in the others category, who reported daytime dysfunction 1 to 2 times a week and/or reasonable.
The participants obtained, on average, a total score of 8.8 points on the PQSI, ranging
from 1 to 21 points. Nursing technicians/assistants had the highest average, and physicians
had the lowest average. Participants of all categories were mostly classified as poor sleepers (74%), especially nursing technicians/assistants (86%). A descriptive analysis
of the PSQI results, according to professional occupation, is shown in [Table 3].
Table 3
PSQI-Br results. Data expressed as absolute and relative frequencies (n [%]). Brazil,
2021 to 2022.
|
PSQI-Br components
|
Total
(n = 125)
|
Dentists
(n = 4)
|
Physiotherapists (n = 26)
|
Nurses
(n = 45)
|
Technicians/ assistants (n = 28)
|
Physicians (n = 8)
|
Others
(n = 14)
|
|
Subjective quality
|
|
|
|
|
|
|
Very good
|
13 (10.4)
|
−
|
4 (15.4)
|
4 (8.9)
|
3 (10.7)
|
−
|
2 (14.3)
|
|
Good
|
60 (48.0)
|
2 (50.0)
|
14 (53.8)
|
18 (40.0)
|
10 (35.7)
|
6 (75.0)
|
10 (71.4)
|
|
Bad
|
38 (30.4)
|
1 (25.0)
|
6 (23.1)
|
17 (37.8)
|
11 (39.3)
|
2 (25.0)
|
1 (7.1)
|
|
Very bad
|
14 (11.2)
|
1 (25.0)
|
2 (7.7)
|
6 (13.3)
|
4 (14.3)
|
−
|
1 (7.1)
|
|
Latency*
|
|
|
|
|
|
|
|
|
≤ 15 min and/or not at all in past month
|
16 (12.8)
|
−
|
1 (3.8)
|
5 (11.1)
|
3 (10.7)
|
3 (37.5)
|
4 (28.6)
|
|
16-30 min and/or < 1 time/week
|
49 (39.2)
|
2 (50.0)
|
12 (46.2)
|
15 (33.3)
|
10 (35.7)
|
5 (62.5)
|
5 (35.7)
|
|
31-60 min and/or < 2–3 times/week
|
29 (23.2)
|
1 (25.0)
|
4 (15.4)
|
16 (35.6)
|
5 (17.9)
|
−
|
3 (21.4)
|
|
> 60 min and/or ≥ 3 times/week
|
31 (24.8)
|
1 (25.0)
|
9 (34.6)
|
9 (20.0)
|
10 (35.7)
|
−
|
2 (14.3)
|
|
Sleep duration
|
|
|
|
|
|
|
> 7 h
|
33 (26.4)
|
2 (50.0)
|
4 (15.4)
|
12 (26.7)
|
7 (25.0)
|
2 (25.0)
|
6 (42.9)
|
|
Between 6 and 7 h
|
35 (28.0)
|
−
|
12 (46.2)
|
8 (17.8)
|
9 (32.1)
|
2 (25.0)
|
4 (28.6)
|
|
Between 5 and 6 h
|
33 (26.4)
|
2 (50.0)
|
6 (23.1)
|
16 (35.6)
|
4 (14.3)
|
1 (12.5)
|
4 (28.6)
|
|
≤ 5 h
|
24 (19.2)
|
−
|
4 (15.4)
|
9 (20.0)
|
8 (28.6)
|
3 (37.5)
|
−
|
|
Sleep efficiency
|
|
|
|
|
|
|
> 85%
|
77 (61.6)
|
2 (50.0)
|
17 (65.4)
|
24 (53.3)
|
17 (60.7)
|
5 (62.5)
|
12 (85.7)
|
|
75–84%
|
24 (19.2)
|
1 (25.0)
|
4 (15.4)
|
10 (22.2)
|
4 (14.3)
|
3 (37.5)
|
2 (14.3)
|
|
65–74%
|
10 (8.0)
|
1 (25.0)
|
1 (3.8)
|
5 (11.1)
|
3 (10.7)
|
−
|
−
|
|
< 65%
|
14 (11.2)
|
−
|
4 (15.4)
|
6 (13.3)
|
4 (14.3)
|
−
|
−
|
|
Sleep disorders
|
|
|
|
|
|
|
None in past month
|
1 (0.8)
|
−
|
−
|
1 (2.2)
|
−
|
−
|
−
|
|
< 1 time/week
|
55 (44.0)
|
2 (50.0)
|
17 (65.4)
|
16 (35.6)
|
7 (25.0)
|
5 (62.5)
|
8 (57.1)
|
|
1–2 times/week
|
57 (45.6)
|
2 (50.0)
|
6 (23.1)
|
25 (55.6)
|
17 (60.7)
|
2 (25.0)
|
5 (35.7)
|
|
≥ 3 times/week
|
12 (9.6)
|
−
|
3 (11.5)
|
3 (6.7)
|
4 (14.3)
|
1 (12.5)
|
1 (7.1)
|
|
Use of sleeping medications
|
|
|
|
|
|
|
Not at all
|
82 (65.6)
|
3 (75.0)
|
16 (61.5)
|
32 (71.1)
|
15 (53.6)
|
7 (87.5)
|
9 (64.3)
|
|
< 1 time/week
|
17 (13.6)
|
−
|
3 (11.5)
|
4 (8.9)
|
5 (17.9)
|
1 (12.5)
|
4 (28.6)
|
|
1–2 times/week
|
9 (7.2)
|
1 (25.0)
|
2 (7.7)
|
3 (6.7)
|
3 (10.7)
|
−
|
−
|
|
≥ 3 times/week
|
17 (13.6)
|
−
|
5 (19.2)
|
6 (13.3)
|
5 (17.9)
|
−
|
1 (7.1)
|
|
Daytime dysfunction†
|
|
|
|
|
|
|
Not once in past month and no difficulties
|
17 (13.6)
|
−
|
7 (26.9)
|
5 (11.1)
|
3 (10.7)
|
−
|
2 (14.3)
|
|
< 1 time/week and/or mild problem
|
55 (44.0)
|
4 (100.0)
|
14 (53.8)
|
19 (53.8)
|
9 (32.1)
|
6 (75.0)
|
3 (21.4)
|
|
1–2 times/week and/or reasonable problem
|
42 (33.6)
|
−
|
1 (3.8)
|
1 (3.8)
|
13 (46.4)
|
2 (25.0)
|
8 (57.1)
|
|
≥ 3 times/week and/or major problem
|
11 (8.8)
|
−
|
4 (15.4)
|
4 (15.4)
|
3 (10.7)
|
−
|
1 (7.1)
|
|
Total score
|
8.8 (4.1)‡
|
8.2 (4.3)‡
|
8.5 (4.3)‡
|
9.2 (4.2)‡
|
10.0 (4.6)‡
|
6.7 (2.6)‡
|
6.9 (2.4)‡
|
|
Classification
|
|
|
|
|
|
|
|
|
Good sleeper
|
32 (25.6)
|
1 (25.0)
|
7 (26.9)
|
12 (26.7)
|
4 (14.3)
|
3 (37.5)
|
5 (35.7)
|
|
Poor sleeper
|
93 (74.4)
|
3 (75.0)
|
19 (73.1)
|
33 (73.3)
|
24 (85.7)
|
5 (62.5)
|
9 (64.3)
|
*Time taken to fall asleep/times when respondent could not fall asleep within 30 min;
†Difficulty staying awake during daily activities and problem maintaining enthusiasm
for usual activities; ‡Mean (SD).
The analysis of psychosocial work aspects showed that the factors in the risk zone for most workers were burnout, stress, and emotional demands. The factors rated as no risk for most workers were quantitative demands, possibilities for development, meaning
of work, commitment to work, recognition, trust in management, justice, role clarity,
social support, job satisfaction, health and wellbeing, and offensive behaviors. Work
pace and work-family conflicts were also factors in the risk zone in most professional'
categories, except for dentists and others (work pace) and dentists and physiotherapists
(work-family conflicts). Work influence and quality of leadership were factors in
the no risk zone in all categories, except nursing technicians/assistants and dentists ([Table 4]).
Table 4
COPSOQ II-BR results. Data expressed as absolute and relative frequencies [n (%)].
Brazil, 2021 to 2022.
|
COPSOQ II-BR Dimensions
|
Total (n = 125)
|
Dentists (n = 4)
|
Physiotherapists (n = 26)
|
Nurses (n = 45)
|
Technicians/ assistants (n = 28)
|
Physicians
(n = 8)
|
Others (n = 14)
|
|
1. Quantitative demands
|
|
|
|
|
|
|
No risk
|
113 (90.4)
|
4 (100.0)
|
23 (88.5)
|
41 (91.1)
|
27 (96.4)
|
6 (75.0)
|
12 (85.7)
|
|
With risk
|
12 (9.6)
|
−
|
3 (11.5)
|
4 (8.9)
|
1 (3.6)
|
2 (25.0)
|
2 (14.3)
|
|
2. Work pace
|
|
|
|
|
|
|
|
|
No risk
|
49 (39.2)
|
3 (75.0)
|
10 (38.5)
|
14 (31.1)
|
11 (39.3)
|
3 (37.5)
|
8 (57.1)
|
|
With risk
|
76 (60.8)
|
1 (25.0)
|
16 (61.5)
|
31 (68.9)
|
17 (60.7)
|
5 (62.5)
|
6 (42.9)
|
|
3. Emotional demands
|
|
|
|
|
|
|
No risk
|
31 (24.8)
|
1 (25.0)
|
9 (34.6)
|
7 (15.6)
|
8 (28.6)
|
1 (12.5)
|
5 (35.7)
|
|
With risk
|
94 (75.2)
|
3 (75.0)
|
17 (65.4)
|
38 (84.4)
|
20 (71.4)
|
7 (87.5)
|
9 (64.3)
|
|
4. Influence at work
|
|
|
|
|
|
|
No risk
|
86 (68.8)
|
3 (75.0)
|
20 (76.9)
|
35 (77.8)
|
11 (39.3)
|
8 (100.0)
|
9 (64.3)
|
|
With risk
|
39 (31.2)
|
1 (25.0)
|
6 (23.1)
|
10 (22.2)
|
17 (60.7)
|
−
|
5 (35.7)
|
|
5. Possibilities for development
|
|
|
|
|
|
|
No risk
|
119 (95.2)
|
3 (75.0)
|
25 (96.2)
|
43 (95.6)
|
27 (96.4)
|
8 (100.0)
|
13 (92.9)
|
|
With risk
|
6 (4.8)
|
1 (25.0)
|
1 (3.8)
|
2 (4.4)
|
1 (3.6)
|
−
|
1 (7.1)
|
|
6. Meaning of work
|
|
|
|
|
|
|
No risk
|
118 (94.4)
|
3 (75.0)
|
25 (96.2)
|
40 (88.9)
|
28 (100.0)
|
8 (100.0)
|
14 (100.0)
|
|
With risk
|
7 (5.6)
|
1 (25.0)
|
1 (3.8)
|
5 (11.1)
|
−
|
−
|
−
|
|
7. Commitment to work
|
|
|
|
|
|
|
No risk
|
115 (92.0)
|
4 (100.0)
|
25 (96.2)
|
41 (91.1)
|
27 (96.4)
|
8 (100.0)
|
10 (71.4)
|
|
With risk
|
10 (8.0)
|
−
|
1 (3.8)
|
4 (8.9)
|
1 (3.6)
|
−
|
4 (28.6)
|
|
8. Predictability
|
|
|
|
|
|
|
|
|
No risk
|
66 (52.8)
|
1 (25.0)
|
16 (61.5)
|
25 (55.6)
|
15 (53.6)
|
5 (62.5)
|
4 (28.6)
|
|
With risk
|
59 (47.2)
|
3 (75.0)
|
10 (38.5)
|
20 (44.4)
|
13 (46.4)
|
3 (37.5)
|
10 (71.4)
|
|
9. Recognition
|
|
|
|
|
|
|
|
|
No risk
|
83 (66.4)
|
2 (50.0)
|
19 (73.1)
|
29 (64.4)
|
17 (60.7)
|
7 (87.5)
|
9 (64.3)
|
|
With risk
|
42 (33.6)
|
2 (50.0)
|
7 (26.9)
|
16 (35.6)
|
11 (39.3)
|
1 (12.5)
|
5 (35.7)
|
|
10. Quality of leadership
|
|
|
|
|
|
|
No risk
|
86 (68.8)
|
1 (25.0)
|
20 (76.9)
|
30 (66.7)
|
22 (78.6)
|
5 (62.5)
|
8 (57.1)
|
|
With risk
|
39 (31.2)
|
3 (75.0)
|
6 (23.1)
|
15 (33.3)
|
6 (21.4)
|
3 (37.5)
|
6 (42.9)
|
|
11. Trust regarding management
|
|
|
|
|
|
|
No risk
|
108 (86.4)
|
3 (75.0)
|
24 (92.3)
|
39 (86.7)
|
25 (89.3)
|
6 (75.0)
|
11 (78.6)
|
|
With risk
|
17 (13.6)
|
1 (25.0)
|
2 (7.7)
|
6 (13.3)
|
3 (10.7)
|
2 (25.0)
|
3 (21.4)
|
|
12. Justice
|
|
|
|
|
|
|
|
|
No risk
|
86 (68.8)
|
4 (100.0)
|
21 (80.8)
|
26 (57.8)
|
19 (67.9)
|
7 (87.5)
|
9 (64.3)
|
|
With risk
|
39 (31.2)
|
−
|
5 (19.2)
|
19 (42.2)
|
9 (32.1)
|
1 (12.5)
|
5 (35.7)
|
|
13. Role clarity
|
|
|
|
|
|
|
|
|
No risk
|
112 (89.6)
|
2 (50.0)
|
26 (100.0)
|
41 (91.1)
|
25 (89.3)
|
7 (87.5)
|
11 (78.6)
|
|
With risk
|
13 (10.4)
|
2 (50.0)
|
−
|
4 (8.9)
|
3 (10.7)
|
1 (12.5)
|
3 (21.4)
|
|
14. Social support
|
|
|
|
|
|
|
|
|
No risk
|
92 (73.6)
|
2 (50.0)
|
19 (73.1)
|
33 (73.3)
|
20 (71.4)
|
7 (87.5)
|
11 (78.6)
|
|
With risk
|
33 (26.4)
|
2 (50.0)
|
7 (26.9)
|
12 (26.7)
|
8 (28.6)
|
1 (12.5)
|
3 (21.4)
|
|
15. Job satisfaction
|
|
|
|
|
|
|
No risk
|
99 (79.2)
|
4 (100.0)
|
22 (84.6)
|
35 (77.8)
|
20 (71.4)
|
7 (87.5)
|
11 (78.6)
|
|
With risk
|
26 (20.8)
|
−
|
4 (15.4)
|
10 (22.2)
|
8 (28.6)
|
1 (12.5)
|
3 (21.4)
|
|
16. Work-family conflicts
|
|
|
|
|
|
|
No risk
|
56 (44.8)
|
3 (75.0)
|
15 (57.7)
|
17 (37.8)
|
13 (46.4)
|
2 (25.0)
|
6 (42.9)
|
|
With risk
|
69 (55.2)
|
1 (25.0)
|
11 (42.3)
|
28 (62.2)
|
15 (53.6)
|
6 (75.0)
|
8 (57.1)
|
|
17. Self-rated health
|
|
|
|
|
|
|
|
|
No risk
|
106 (84.8)
|
3 (75.0)
|
24 (92.3)
|
37 (82.2)
|
21 (75.0)
|
8 (100.0)
|
13 (92.9)
|
|
With risk
|
19 (15.2)
|
1 (25.0)
|
2 (7.7)
|
8 (17.8)
|
7 (25.0)
|
−
|
1 (7.1)
|
|
18. Burnout
|
|
|
|
|
|
|
|
|
No risk
|
18 (14.4)
|
1 (25.0)
|
5 (19.2)
|
5 (11.1)
|
4 (14.3)
|
1 (12.5)
|
2 (14.3)
|
|
With risk
|
107 (85.6)
|
3 (75.0)
|
21 (80.8)
|
40 (88.9)
|
24 (85.7)
|
7 (87.5)
|
12 (85.7)
|
|
19. Stress
|
|
|
|
|
|
|
|
|
No risk
|
24 (19.2)
|
1 (25.0)
|
7 (26.9)
|
7 (15.6)
|
7 (25.0)
|
1 (12.5)
|
1 (7.1)
|
|
With risk
|
101 (80.8)
|
3 (75.0)
|
19 (73.1)
|
38 (84.4)
|
21 (75.0)
|
7 (87.5)
|
13 (92.9)
|
|
20. Unwanted sexual attention
|
|
|
|
|
|
|
No risk
|
106 (84.8)
|
3 (75.0)
|
24 (92.3)
|
34 (75.6)
|
25 (89.3)
|
7 (87.5)
|
13 (92.9)
|
|
Risk
|
19 (15.2)
|
1 (25.0)
|
2 (7.7)
|
11 (24.4)
|
3 (10.7)
|
1 (12.5)
|
1 (7.1)
|
|
21. Threats of violence
|
|
|
|
|
|
|
No risk
|
93 (74.4)
|
3 (75.0)
|
25 (96.2)
|
30 (66.7)
|
18 (64.3)
|
5 (62.5)
|
12 (85.7)
|
|
Risk
|
32 (25.6)
|
1 (25.0)
|
1 (3.8)
|
15 (33.3)
|
10 (35.7)
|
3 (37.5)
|
2 (14.3)
|
|
22. Physical violence
|
|
|
|
|
|
|
|
|
No risk
|
114 (91.2)
|
4 (100.0)
|
25 (96.2)
|
42 (93.3)
|
22 (78.6)
|
8 (100.0)
|
13 (92.9)
|
|
Risk
|
11 (8.8)
|
−
|
1 (3.8)
|
3 (6.7)
|
6 (21.4)
|
−
|
1 (7.1)
|
|
23. Bullying
|
|
|
|
|
|
|
|
|
No risk
|
104 (83.2)
|
3 (75.0)
|
22 (84.6)
|
38 (84.4)
|
22 (78.6)
|
7 (87.5)
|
12 (85.7)
|
|
Risk
|
21 (16.8)
|
1 (25.0)
|
4 (15.4)
|
7 (15.6)
|
6 (21.4)
|
1 (12.5)
|
2 (14.3)
|
Abbreviation: COPSOQ II-BR, Copenhagen Psychosocial Questionnaire II validated for
Brazilian Portuguese.
Significant correlations ranging from weak to moderate were found between sleep quality
and the following variables: work pace, predictability, justice, work-family conflict,
self-rated health, burnout, and stress ([Table 5]). Nonsignificant correlations (P > 0.05) are not shown.
Table 5
Significant correlations (P < 0.05) between sleep quality and psychosocial factors.
|
Psychosocial factors
|
rpb
|
Interpretation†
|
|
Work pace
|
0.24
|
Weak
|
|
Predictability
|
0.30
|
Moderate
|
|
Justice
|
0.20
|
Weak
|
|
Work family conflicts
|
0.28
|
Weak
|
|
Self-rated health
|
0.20
|
Weak
|
|
Burnout
|
0.33
|
Moderate
|
|
Stress
|
0.18
|
Weak
|
*The sleep quality refers to the score of PSQI-Br and positive correlations means
that psychosocial risks were directly correlated with poorer sleep quality.
The proportion of good and poor sleepers in each risk zone is shown in [Figure 1]. The proportion of poor sleepers ranged from 78.2% for stress to 94.7% for self-rated health in the risk zone. In contrast, the proportion of good sleepers in the no risk zone ranged from 29.2% for self-rated health to 61.1% for burnout. Furthermore, the
proportion of poor sleepers was higher in both risk categories in the psychosocial factors presented,
except for the no risk zone of burnout, where the majority were classified as good sleepers.
Fig. 1 Distribution of workers according to quality of sleep for work pace, predictability,
justice, work-family conflict, self-rated health, burnout, and stress.
Discussion
The study was conducted with frontline public healthcare workers during the COVID-19
pandemic, who were deeply affected by precarious working conditions and by the intensification
of preexisting psychosocial factors at work, which impacted their health conditions,
including sleep quality. Results showed that the psychosocial factors in the risk zone that most impact sleep quality were burnout and predictability. Moreover, work pace,
justice, work-family conflicts, self-rated health, and stress had weak significant
correlations with sleep quality.
Approximately 10% of the workers reported being service providers (8.8%) or outsourced
(1.6%), demonstrating a trend toward the outsourcing of labor in the public healthcare
system. This trend was described as predominant in a previous survey, which also reported
increasing difficulties for medical professionals to find jobs with a formal contract.[31] Thus, there is as association between the more flexible nature of the work relationship
and both the increase in uncontrolled working hours and the number of employment ties
assumed by workers, culminating in the further precariousness of working conditions
as well as the loss of labor rights, social security rights, and protection from various
risks.[31]
A large portion of workers reported having more than one job (30%), which suggests
the search for solutions for the devaluation of wages or professional dissatisfaction.
Multiple jobs are one of the main factors that cause stress, and this problem is even
more evident in female professionals, who have an additional workload consisting of
domestic chores and children.[32]
Most participants were classified as poor sleepers (74.4%), which is similar to findings described in previous studies,[33]
[34]
[35]
[36] with small differences regarding the prevalence of poor sleep quality in comparison
to some surveys. This difference may be explained by the professional categories in
the sample analyzed. A Brazilian study found a predominance of poor sleep quality
among nurses (72%) and nursing technicians (88%).[36] Another study conducted with nurses in Ethiopia found that 75% had poor sleep quality.[35] In the present investigation, poor sleep quality was also found among nursing staff,
with the worse rate among nursing technicians/assistants (86%).
A study conducted with healthcare workers in the Middle East during the COVID-19 pandemic
found that 75% reported poor sleep quality; physicians had the lowest mean total PSQI
score (6.6 points), followed by nursing staff (7.0 points) and other healthcare professionals
(7.8 points).[34] In the present study, physicians also had the lowest mean total PSQI score (6.7
points). However, the results regarding the nursing staff differ from data found in
the literature, especially in the category of nursing technicians/assistants, who
had the highest mean (10 points).
Poor sleep quality is common among nursing staff, with a combined prevalence of 61%
in a meta-analysis conducted before the COVID-19 pandemic, and a mean total PSQI score
of 7 points.[37] A Chinese study also found a greater frequency of poor sleep quality among nursing
staff compared to other healthcare workers.[38]
The main psychosocial factors in the risk zone among the participants were emotional demands, burnout, and stress. The high prevalence
of these factors suggests tense, demanding, stressful work environments. These results
are similar to findings reported in a Chinese study, which found stress, quantitative
demands, and burnout were the main risk factors among healthcare workers.[39]
Psychosocial factors became more evident with the COVID-19 pandemic, which put pressure
on healthcare systems and, consequently, their workers,[40] driving changes in work organization to deal with the increased number of patients,
care demands as well as tension and stress among workers.[41] A literature review pointed out that the main problem found in frontline healthcare
teams, especially among nursing staff, was anxiety, followed by depression, stress,
and sleep disturbances.[42]
Moderate correlations were found between sleep quality and both burnout and predictability.
Thus, workers with burnout and low predictability at work have worse sleep quality.
Poor sleep quality, especially the presence of daytime dysfunction and working long
shifts, worsens the symptoms of burnout,[9] since there is no adequate recovery, generating a vicious cycle that is harmful
to health.
Work pace, which was a psychosocial factor in the risk zone, especially among nurses, physicians, and nursing technicians/assistants (69%, 62%,
and 61%, respectively), was significantly associated with sleep quality, with a higher
number of poor sleepers in the risk zone for this factor (83%). A study conducted during the COVID-19 pandemic also found
an association between poor sleep quality and feeling moderate/heavy overwork.[43] Furthermore, work overload, which is the result of insufficient number of professionals
and a lack of organizational support, is common in nursing work,[44] which may explain the greater effect on sleep quality described in the literature[37]
[38] and found in the present study.
Purposive sampling and sample size are important limitations of the present study.
Therefore, this study was composed of a convenience sample, which affects the generalizability
of our findings. The pandemic context may have made participation in our study difficult,
since most healthcare workers had high work demands and did not have time to participate
in our study. Furthermore, the online form was very extensive, which may have also
made it difficult for workers to participate. Also, the online design of the study
may limit the participation of workers less familiar with electronic resources, given
the greater participation of younger workers with higher education levels. Future
studies in this field could recruit a larger sample size, which enables the use of
regression models.
Despite the limitations, the present findings enable reflections on the occurrence
of psychosocial factors in healthcare work and the poor quality of sleep among these
workers. Moreover, the results reveal the correlation between these variables considering
the particularities of each professional category and the context of the pandemic,
which aggravated the precarious working conditions of these workers in Brazil, accentuating
the existing weaknesses in healthcare services, the effects of which need to be investigated
in the long term in this population.
Conclusions
Our findings showed that psychosocial factors in the risk zone are prevalent among healthcare workers, especially burnout, stress, and emotional
demands; they also showed that most workers had poor sleep quality, especially the
nursing staff, and that there is a moderate correlation between sleep quality and
both burnout and predictability.
Therefore, it is important to recognize the health risk factors at work and intervene
to mitigate or eliminate disadvantage factors with a view to protecting the health
of workers. Health institutions should commit to providing better working conditions
to minimize stress and unpredictability and to address the mental and physical suffering
of workers, seeking effective strategies to improve their sleep quality as well as
their personal and professional wellbeing, which are closely related. Also, future
studies should investigate the impacts of psychosocial factors at work on the sleep
quality and health of workers in the long term.
