Keywords
sleep quality - sleep disorders - shift work - medication use
Introduction
Sleep is a physiological and biological process which is fundamental for human survival.[1] It performs a multiplicity of vital functions, such as the conservation and restoration
of energy, the regulation of metabolic processes, among others.[2]
[3]
[4] Disturbances in sleep quality can trigger significant changes in the individual's
physical, occupational, cognitive and social functioning, substantially compromising
quality of life.[2] Low quality and/or quantity of sleep is considered a public health problem.[5] Sleep disorders are associated, among other pathologies, with an increased risk
of acute myocardial infarction, stroke, and depression.[5]
Sleep quality can be influenced by different factors, such as previous medical conditions,
use of medications and/or stimulant substances, working schedules, among others.[3]
In the labor context, there is an increasing number of workers performing their tasks
in shifts in several activity sectors, as a consequence of technological changes and
economic globalization, which require the availability of goods and services 24/7.[6]
[7] Those work conditions may enhance changes in the individual's endogenous biological
rhythm, resulting in a temporal conflict between the biological clock and the externally-imposed
social scheme.[8] The use of medication to relieve sleep disorders caused by working hours, such as
insomnia, is common; however, if abused, without supervision and monitoring, it can
trigger new pathological conditions.[9]
[10] On the other hand, the prescription of different medications for the treatment of
several clinical conditions can have adverse effects in sleep.[10]
Therefore, it is of paramount importance to evaluate medication use in shift workers
compared to daytime workers and its consequences on sleep quality, to promote the
rational and safe use of medications.
The present work aims to determine the association between medication use and sleep
quality in shift workers versus daytime workers.
Materials and Methods
Study Design and Study Population
We conducted a quantitative cross-sectional study using a convenience sample of active
workers, regardless of the activity sector. The participants were divided in two groups:
shift and daytime schedule workers. Individuals aged ≥ 18 years who had been holding
a job for at least six months were included.
Data Collection
Data was collected using a combination of three questionnaires: the first one, which
was developed for the present study to collect sociodemographic characteristics, associated
comorbidities and medication use; the Epworth Sleepiness Scale (ESS) was used to assess
the level of sleepiness; and the Pittsburgh Sleep Quality Index (PSQI), to assess
the sleep quality. The questionnaires were available online between October and December
2021, and the link to access them was sent through e-mail and social media using the
“snowball” method. Access to the questionnaires was only granted after agreement with
a participant information document, which contained all the necessary information
for participation in the study.
Data Analyses
The IBM SPSS Statistics for Windows (IBM Corp., Armonk, NY, United States) software,
version 26.0, and the Microsoft Excel software (Microsoft Corp., Redmond, WA, United
States) were used for the statistical analysis. Simple descriptive analysis (absolute
and relative frequencies) was performed for the nominal and ordinal variables, as
well as central tendency and dispersion measures for the quantitative variables. Bivariate
associations between variables were determined using the Chi-squared statistical test,
with a significance level of 5%. A logistic regression model was developed, in which
all variables that showed an association with sleep quality were included, also adopting
a significance level of 5%.
Ethical Considerations
The present study was approved by the Ethics Committee of Escola Superior de Tecnologia
da Saúde de Lisboa under process number 41-2021. Anonymity and confidentiality of
the data obtained were guaranteed. All participants could withdraw from the study,
hide any information, or refuse to answer any question.
Results
Characterization of Study Participants
Of the 296 study participants, 109 individuals (36.82%) were male and 187 (63.18%),
female, with a mean age of 36.03 ± 8.93 years, and 130 (43.92%) subjects worked in
the health sector. The mean time of professional activity was of 11 ± 8.69 years.
Most (58.11%) of the participants worked in shifts, and 51.7% had been working in
their current shifts for five or more years.
No cardiovascular (CV) risk factors were reported by 182 (61.49%) participants; among
those who did report risk factors, smoking had the highest prevalence. In total, 87
(29.39%) participants, mainly women, reported having some prediagnosed pathology,
specially psychological ones.
In terms of sleep disorders, most participants (91.89%) reported having none. Of the
24 participants who reported having a sleep disorder, insomnia was the most mentioned.
All data are shown in [Table 1].
Table 1
Characteristics of the study participants.
|
Male
|
Female
|
Total
|
|
Study participants
|
109 (36.82%)
|
187 (63.18%)
|
296 (100.00%)
|
|
Age (in years): mean ± standard deviation
|
38.50 ± 9.69
|
34.59 ± 8.12
|
36.03 ± 8.93
|
|
Professional sector: n (%)
|
|
|
|
|
Health
|
27 (24.77%)
|
103 (55.08%)
|
130 (43.92%)
|
|
Industry
|
66 (60.55%)
|
50 (26.74%)
|
116 (39.19%)
|
|
Other
|
16 (14.68%)
|
34 (18.18%)
|
50 (16.89%)
|
|
Time of professional activity (in years): mean ± standard deviation
|
12 ± 9.96
|
10 ± 7.72
|
11 ± 8.69
|
|
Type of work: n (%)
|
|
|
|
|
Day work
|
23 (21.10%)
|
101 (54.01%)
|
124 (41.89%)
|
|
Shift work
|
86 (78.90%)
|
86 (45.99%)
|
172 (58.11%)
|
|
Current shift time (in years): n (%)
|
|
|
|
|
< 5
|
52 (47.7%)
|
91 (48.7%)
|
143 (48.3%)
|
|
≥ 5
|
57 (52.3%)
|
96 (51.3%)
|
153 (51.7%)
|
|
Known cardiovascular risk factor: n (%)
|
|
|
|
|
Yes
|
51 (46.79%)
|
63 (33.69%)
|
114 (38.51%)
|
|
No
|
58 (53.21%)
|
124 (66.31%)
|
182 (61.49%)
|
|
Type of cardiovascular risk factor (n = 114): n (%)
|
|
|
|
|
Diabetes
|
4 (7.84%)
|
2 (3.17%)
|
6 (5.26%)
|
|
Smoking
|
29 (56.86%)
|
36 (57.14%)
|
65 (57.02%)
|
|
Hypertension
|
16 (31.37%)
|
11 (17.46%)
|
27 (23.68%)
|
|
Dyslipidemia
|
2 (3.92%)
|
4 (6.35%)
|
6 (5.26%)
|
|
Obesity
|
13 (25.49%)
|
18 (28.57%)
|
31 (27.19%)
|
|
Other
|
22 (43,14%)
|
30 (47.62%)
|
52 (45.61%)
|
|
Diagnosed pathology (n = 87): n (%)
|
|
|
|
|
Cardiovascular
|
4 (4.60%)
|
4 (4.60%)
|
8 (9.20%)
|
|
Psychological
|
24 (27.60%)
|
43 (49.43%)
|
67 (77.01%)
|
|
Other
|
2 (2.30%)
|
10 (11.49%)
|
12 (13.79%)
|
|
Sleep disorder: n (%)
|
|
|
|
|
Yes
|
11 (10.09%)
|
13 (6.95%)
|
24 (8.11%)
|
|
No
|
98 (89.91%)
|
10 (76.92%)
|
272 (91.89%)
|
|
Type of sleep disorder (n = 24): n (%)
|
|
|
|
|
Insomnia
|
7 (63.64%)
|
10 (76.92%)
|
17 (70.83%)
|
|
Restless legs syndrome
|
2 (18.18%)
|
2 (15.38%)
|
4 (16.67%)
|
|
Excessive sleepiness
|
5 (45.45%)
|
6 (46.15%)
|
4 (16.67%)
|
|
Obstructive sleep apnea syndrome
|
4 (36.36%)
|
2 (15.38%)
|
11 (45.83%)
|
|
Parasomnias
|
1 (9.09%)
|
1 (7.69%)
|
2 (8.33%)
|
Medication Use
Daily use of medications was reported by 87 (29.39%) subjects and it was slightly
higher among female participants (32.09%). Concerning the person who recommended the
use of the last medication, there was an expressive value referring to self-initiative
(41.22%), even though physician's recommendation had the higher frequency (50.34%).
Regarding the last drug used, there were no considerable differences in terms of the
frequencies of drugs acting or not in the central nervous system (CNS). Most drugs
used had no proven effects on sleep (72.30%), and 66.22% of the study participants
stated that they did not take sleep medications ([Table 2]).
Table 2
Characterization of medication use.
|
Male: n (%)
|
Female n: (%)
|
Total: n (%)
|
|
Medication use
|
|
|
|
|
Daily
|
27 (24.77%)
|
60 (32.09%)
|
87 (29.39%)
|
|
Daily (contraceptive)
|
0 (0.00%)
|
6 (3.21%)
|
6 (2.03%)
|
|
Sporadic
|
82 (75.23%)
|
121 (64.71%)
|
203 (68.58%)
|
|
Referring agent
|
|
|
|
|
Family member
|
2 (1.83%)
|
0 (0.00%)
|
2 (0.68%)
|
|
Pharmacy professional
|
8 (7.34%)
|
11 (5.88%)
|
19 (6.42%)
|
|
Self-initiative
|
41 (37.61%)
|
81 (43.32%)
|
122 (41.22%)
|
|
Physician
|
55 (50.46%)
|
94 (50.27%)
|
149 (50.34%)
|
|
Friend
|
2 (1.83%)
|
0 (0.00%)
|
2 (0.68%)
|
|
Other
|
1 (0.92%)
|
1 (0.53%)
|
2 (0.68%)
|
|
Direct action in the central nervous system
|
|
|
|
|
Yes
|
41 (37.62%)
|
77 (41.18%)
|
118 (39.87%)
|
|
No
|
49 (44.95%)
|
93 (49.73%)
|
142 (47.97%)
|
|
Not applicable
|
19 (17.43%)
|
17 (9.09%)
|
36 (12.16%)
|
|
Proven effects on sleep
|
|
|
|
|
Yes
|
16 (14.68%)
|
30 (16.04%)
|
46 (15.54%)
|
|
No
|
74 (67.89%)
|
140 (74.87%)
|
214 (72.30%)
|
|
Not applicable
|
19 (17.43%)
|
17 (9.09%)
|
36 (12.16%)
|
|
Sleep medication
|
|
|
|
|
Yes
|
41 (37.61%)
|
59 (31.55%)
|
100 (33.78%)
|
|
No
|
68 (62.39%)
|
128 (68.45%)
|
196 (66.22%)
|
A higher use of CNS-acting drugs was found among respondents with sleep disorders,
but it was not statistically significant (p = 0.063). Regarding drugs with proven effects on sleep, there was an association
between their use and the presence of sleep disorders (p < 0.001) ([Tables 3A] and [3B]).
Table 3A
Use of drugs acting in the central nervous system.
|
Direct action in the central nervous system
|
|
Yes: n (%)
|
No: n (%)
|
Not applicable: n (%)
|
p-value
|
|
Gender
|
|
|
|
|
|
Male
|
41 (37.6%)
|
49 (45.0%)
|
19 (17.4%)
|
0.106
|
|
Female
|
77 (41.2%)
|
93 (49.7%)
|
17 (9.1%)
|
|
|
Variables related to professional activity
|
|
Sector of activity
|
|
|
|
|
|
Health
|
53 (40.8%)
|
67 (51.5%)
|
10 (7.7%)
|
|
|
Industry
|
41 (35.3%)
|
57 (49.1%)
|
18 (15.5%)
|
0.129
|
|
Other
|
24 (48.0%)
|
18 (36.0%)
|
8 (16.0%)
|
|
|
Time of professional activity (in years)
|
|
|
|
|
|
< 5
|
48 (43.6%)
|
46 (41.8%)
|
16 (14.5%)
|
0.244
|
|
≥ 5
|
70 (37.6%)
|
96 (51.6%)
|
20 (10.8%)
|
|
|
Current type of work
|
|
|
|
|
|
Day work
|
53 (42.7%)
|
61 (49.2%)
|
10 (8.1%)
|
0.178
|
|
Shift work
|
65 (37.8%)
|
81 (47.1%)
|
26 (15.1%)
|
|
|
Time in the current work
|
|
|
|
|
|
≤ 5 years
|
61 (42.7%)
|
67 (46.9%)
|
15 (10.5%)
|
0.535
|
|
> 5 years
|
57 (37.3%)
|
75 (49.0%)
|
21 (13.7%)
|
|
|
Variables related to comorbidities
|
|
Cardiovascular risk factor
|
|
|
|
|
|
Yes
|
38 (33.3%)
|
59 (51.8%)
|
17 (14.9%)
|
0.158
|
|
No
|
80 (44.0%)
|
83 (45.6%)
|
19 (10.4%)
|
|
|
Sleep disorders
|
|
|
|
|
|
Yes
|
14 (58.3%)
|
6 (25.0%)
|
4 (16.7%)
|
0.063
|
|
No
|
104 (38.2%)
|
136 (50.0%)
|
32 (11.8%)
|
|
|
Epworth Sleepiness Scale
|
|
|
|
|
|
Normal
|
53 (41.7%)
|
56 (44.1%)
|
18 (14.2%)
|
|
|
Moderate sleepiness
|
18 (39.1%)
|
24 (52.2%)
|
4 (8.7%)
|
0.762
|
|
Abnormal sleepiness
|
47 (38.2%)
|
62 (50.4%)
|
14 (11.4%)
|
|
Table 3B
Use of drugs with proven effects on sleep.
|
Drugs with proven effects on sleep
|
|
Yes: n (%)
|
No: n (%)
|
Not available: n (%)*
|
p-value
|
|
Gender
|
|
|
|
|
|
Male
|
16 (14.7%)
|
74 (67.9%)
|
19 (17.4%)
|
0.106
|
|
Female
|
30 (16.0%)
|
140 (74.9%)
|
17 (9.1%)
|
|
|
Variables related to professional activity
|
|
Activity sector
|
|
|
|
|
|
Health
|
23 (17.7%)
|
97 (74.6%)
|
10 (7.7%)
|
0.292
|
|
Industry
|
15 (12.9%)
|
83 (71.6%)
|
18 (15.5%)
|
|
|
Other
|
8 (16.0%)
|
34 (68.0%)
|
8 (16.0%)
|
|
|
Time of professional activity
|
|
|
|
|
|
< 5 years
|
13 (11.8%)
|
81 (73.6%)
|
16 (14.5%)
|
0.298
|
|
≥ 5 years
|
33 (17.7%)
|
133 (71.5%)
|
20 (10.8%)
|
|
|
Current work
|
|
|
|
|
|
Day wok
|
20 (16.1%)
|
94 (75.8%)
|
10 (8.1%)
|
0.187
|
|
Shift work
|
26 (15.1%)
|
120 (69.8%)
|
26 (15.1%)
|
|
|
Time in the current work
|
|
|
|
|
|
< 5 years
|
21 (14.7%)
|
107 (74.8%)
|
15 (10.5%)
|
0.603
|
|
≥ 5 years
|
25 (16.3%)
|
107 (69.9%)
|
21 (13.7%)
|
|
|
Variables related to comorbidities
|
|
Cardiovascular risk factor
|
|
|
|
|
|
Yes
|
18 (15.8%)
|
79 (69.3%)
|
17 (14.9%)
|
|
|
No
|
28 (15.4%)
|
135 (74.2%)
|
19 (10.4%)
|
0.499
|
|
Sleep disorders
|
|
|
|
|
|
Yes
|
14 (58.3%)
|
6 (25.0%)
|
4 (16.7%)
|
< 0.001
|
|
No
|
32 (11.8%)
|
208 (76.5%)
|
32 (11.8%)
|
|
|
Epworth Sleepiness Scale
|
|
|
|
|
|
Normal
|
22 (17.3%)
|
87 (68.5%)
|
18 (14.2%)
|
|
|
Moderate sleepiness
|
3 (6.5%)
|
39 (84.8%)
|
4 (8.7%)
|
0.283
|
|
Excessive sleepiness
|
21 (17.1%)
|
88 (71.5%)
|
14 (11.4%)
|
|
Note: *Refers to named substances that are not actually considered medicines, such
as multivitamins, for example.
Sleep Quality Assessment
Male participants reported worse sleep quality (p = 0.005), such as shift workers (p < 0.001), and those who had been on the current shift for less than 5 years (p = 0.023).
The presence of CV risk factors was also associated with poor sleep quality (p < 0.001). Participants with a previously-diagnosed phycological pathology reported
worse sleep quality (p < 0.001). All participants with sleep disorders had poor sleep quality (p < 0.001) ([Table 4]).
Table 4
Sleep quality according to the Pittsburgh Sleep Quality Index.
|
Sleep Quality
|
|
Good: n (%)
|
Poor: n (%)
|
p-value
|
|
Gender
|
|
|
|
|
Male
|
31 (28.4%)
|
78 (71.6%)
|
0.005
|
|
Female
|
84 (44.9%)
|
103 (55.1%)
|
|
|
Variables related to professional activity
|
|
Activity sector
|
|
|
|
|
Health
|
70 (53.8%)
|
60 (46.2%)
|
< 0.001
|
|
Industry
|
25 (21.6%)
|
91 (78.4%)
|
|
|
Other
|
20 (40.0%)
|
30 (60.0%)
|
|
|
Time of professional activity
|
|
|
|
|
< 5 years
|
36 (32.7%)
|
74 (67.3%)
|
0.096
|
|
≥ 5 years
|
79 (42.5%)
|
107 (57.5%)
|
|
|
Current work
|
|
|
|
|
Day work
|
70 (56.5%)
|
54 (43.5%)
|
0.001
|
|
Shift work
|
45 (26.2%)
|
127 (73.8%)
|
|
|
Time in the current work
|
|
|
|
|
< 5 years
|
46 (32.2%)
|
97 (67.8%)
|
0.023
|
|
≥ 5 years
|
69 (45.1%)
|
84 (54.9%)
|
|
|
Variables related to comorbidities and preexisting pathologies
|
|
Cardiovascular risk factors
|
|
|
|
|
Yes
|
27 (23.7%)
|
87 (76.3%)
|
< 0.001
|
|
No
|
88 (48.4%)
|
94 (51.6%)
|
|
|
Previously diagnosed pathologies
|
|
|
|
|
Cardiovascular
|
|
|
|
|
Yes
|
2 (25.0%)
|
6 (75.0%)
|
0.415
|
|
No
|
113 (39.2%)
|
175 (60.8%)
|
|
|
Psychological
|
|
|
|
|
Yes
|
13 (19.4%)
|
54 (80.6%)
|
< 0.001
|
|
No
|
102 (44.5%)
|
127 (55.5%)
|
|
|
Other
|
|
|
|
|
Yes
|
4 (33.3%)
|
8 (66.7%)
|
0.689
|
|
No
|
111 (39.1%)
|
173 (60.9%)
|
|
|
Existence of sleep disorders
|
|
|
|
|
Yes
|
0 (0.0%)
|
24 (100.0%)
|
< 0.001
|
|
No
|
115 (42.3%)
|
157 (57.7%)
|
|
|
Type of sleep disorder
|
|
|
|
|
Insomnia
|
|
|
|
|
Yes
|
0 (0.0%)
|
17 (100.0%)
|
Not applicable
|
|
No
|
0 (0.0%)
|
7 (100.0%)
|
|
|
Restless legs syndrome
|
|
|
|
|
Yes
|
0 (0.0%)
|
4 (100.0%)
|
Not applicable
|
|
No
|
0 (0.0%)
|
20 (100.0%)
|
|
|
Excessive sleepiness
|
|
|
|
|
Yes
|
0 (0.0%)
|
11 (100.0%)
|
Not applicable
|
|
No
|
0 (0.0%)
|
13 (100.0%)
|
|
|
Obstructive sleep apnea syndrome
|
|
|
|
|
Yes
|
0 (0.0%)
|
6 (100.0%)
|
Not applicable
|
|
No
|
0 (0.0%)
|
18 (100.0%)
|
|
|
Parasomnias
|
|
|
|
|
Yes
|
0 (0.0%)
|
2 (100.0%)
|
Not applicable
|
|
No
|
0 (0.0%)
|
22 (100.0%)
|
|
When analyzing the association between medication use and sleep quality, no association
was found regarding the daily or sporadic use of drugs and sleep quality. Participants
with insomnia, a chronic disease or a psychological problem had worse sleep quality
(p = 0.010), and the participants who were using sleeping drugs also reported worse
sleep quality (p < 0.001). No association was found between sleep quality and the use of drugs acting
on the CNS, although there was a tendency for worse sleep quality in participants
using drugs with proven effects on sleep (p = 0.054) ([Table 5]).
Table 5
Medication use and sleep quality.
|
Sleep quality
|
|
Good: n (%)
|
Poor: n (%)
|
p-value
|
|
Medication use
|
|
|
|
|
Daily
|
33 (35.5%)
|
60 (64.5%)
|
0.421
|
|
Sporadic
|
82 (40.4%)
|
121 (59.6%)
|
|
|
Reason for last medication
|
|
|
|
|
Unusual pain or symptoms
|
51 (42.1%)
|
70 (57.9%)
|
|
|
Feeling weak
|
7 (36.8%)
|
12 (63.2%)
|
|
|
Psychological problem
|
7 (35.0%)
|
13 (65.0%)
|
|
|
Chronic disease
|
12 (34.3%)
|
23 (65.7%)
|
0.010
|
|
Insomnia
|
1 (4.3%)
|
22 (95.7%)
|
|
|
Health problem
|
18 (40.9%)
|
26 (59.1%)
|
|
|
Other
|
19 (55.9%)
|
15 (44.1%)
|
|
|
Agent who indicated the last medication
|
|
|
|
|
Familiar
|
0 (0.0%)
|
2 (100.0%)
|
|
|
Pharmacy professional
|
4 (21.1%)
|
15 (78.9%)
|
|
|
Self-initiative
|
53 (43.4%)
|
69 (56.6%)
|
0.280
|
|
Medical doctor
|
57 (38.3%)
|
92 (61.7%)
|
|
|
Friend
|
0 (0.0%)
|
2 (100.0%)
|
|
|
Other
|
1 (50.0%)
|
1 (50.0%)
|
|
|
Sleeping medication
|
|
|
|
|
Yes
|
14 (14.0%)
|
86 (86.0%)
|
< 0.001
|
|
No
|
101 (51.5%)
|
95 (48.5%)
|
|
|
Drugs acting in the central nervous system
|
|
|
|
|
Yes
|
49 (41.5%)
|
69 (58.5%)
|
|
|
No
|
57 (40.1%)
|
85 (59.9%)
|
0.186
|
|
Not applicable
|
9 (25.0%)
|
27 (75.0%)
|
|
|
Drugs with proven effects on sleep
|
|
|
|
|
Yes
|
14 (30.4%)
|
32 (69.6%)
|
|
|
No
|
92 (43.0%)
|
122 (57.0%)
|
0.054
|
|
Not applicable
|
9 (25.0%)
|
27 (75.0%)
|
|
The use of stimulant drinks (p = 0.004) and stimulant medication (p = 0.044) were associated with decreasing sleep quality: as the frequency of consumption
increases, sleep quality decreases. Excessive sleepiness was also associated with
worse sleep quality (p < 0.001): the greater the degree of sleepiness, the worse the sleep quality ([Table 6]).
Table 6
Use of substances that can affect sleep, sleepiness, and sleep quality.
|
Sleep quality
|
|
Good: n (%)
|
Poor: n (%)
|
p-value
|
|
Coffee
|
|
|
|
|
Never
|
13 (48.1%)
|
14 (51.9%)
|
|
|
Rarely
|
9 (39.1%)
|
14 (60.9%)
|
0.107
|
|
Sometimes
|
15 (25.4%)
|
44 (74.6%)
|
|
|
Often
|
78 (41.7%)
|
109 (58.3%)
|
|
|
Alcohol
|
|
|
|
|
Never
|
14 (31.1%)
|
31 (68.9%)
|
|
|
Rarely
|
46 (38.0%)
|
75 (62.0%)
|
0.588
|
|
Sometimes
|
48 (42.9%)
|
64 (57.1%)
|
|
|
Often
|
7 (38.9%)
|
11 (61.1%)
|
|
|
Xanthines
|
|
|
|
|
Never
|
12 (36.4%)
|
21 (63.6%)
|
|
|
Rarely
|
36 (42.9%)
|
48 (57.1%)
|
0.797
|
|
Sometimes
|
43 (38.7%)
|
68 (61.3%)
|
|
|
Often
|
24 (35.3%)
|
44 (64.7%)
|
|
|
Stimulant drinks
|
|
|
|
|
Never
|
71 (48.3%)
|
76 (51.7%)
|
|
|
Rarely
|
39 (32.2%)
|
82 (67.8%)
|
0.004
|
|
Sometimes
|
5 (20.8%)
|
19 (79.2%)
|
|
|
Often
|
0 (0.0%)
|
4 (100.0%)
|
|
|
Stimulant medications
|
|
|
|
|
Never
|
78 (44.6%)
|
97 (55.4%)
|
|
|
Rarely
|
35 (33.0%)
|
71 (67.0%)
|
0.044
|
|
Sometimes
|
2 (14.3%)
|
12 (85.7%)
|
|
|
Often
|
0 (0.0%)
|
1 (100.0%)
|
|
|
Epworth Sleepiness Scale
|
|
|
|
|
Normal
|
69 (54.3%)
|
58 (45.7%)
|
0.001
|
|
Moderate sleepiness
|
17 (37.0%)
|
29 (63.0%)
|
|
|
Excessive sleepiness
|
29 (23.6%)
|
94 (76.4%)
|
|
Finally, we developed a logistic regression model to adjust all the variables associated
with sleep quality in the bivariate analysis, in order to determine the risk of having
poor sleep quality. The model shows that, after adjustment, only time on current shift
(p = 0.003) and the degree of excessive sleepiness (p = 0.043) are associated with sleep quality ([Table 7]).
Table 7
Risk factors for poor sleep quality.
|
Ajusted odds ratiob
|
p-value
|
|
Gender
|
|
0.746
|
|
Male
|
1.120 (0.563–2.227)
|
|
|
Female
|
1
|
|
|
Age group
|
|
0.558
|
|
< 45 years old)
|
0.558 (0.220–1.414)
|
|
|
≥ 45 years old)
|
1
|
|
|
Sector of activity
|
|
|
|
Health
|
0.722 (0.298–1.751)
|
0.426
|
|
Industry
|
1.181 (0.417–3.342)
|
0.471
|
|
Other
|
1
|
0.754
|
|
Drugs with proven effect on sleep
|
|
0.954
|
|
No effect
|
0.971 (0.360–2.622)
|
|
|
Effect
|
1
|
|
|
Current work
|
|
0.064
|
|
Day work
|
0.518 (0.258–1.039)
|
|
|
Shift work
|
1
|
|
|
Time in current work
|
|
0.003
|
|
< 5 years
|
2.737 (1.422–5.268)
|
|
|
≥ 5 years
|
1
|
|
|
Cardiovascular risk factor
|
|
0.061
|
|
No
|
0.519 (0.262–1.030)
|
|
|
Yes
|
1
|
|
|
Diagnosed psychological pathology
|
|
0.775
|
|
No
|
0.866 (0.322–2.330)
|
|
|
Yes
|
1
|
|
|
Daily use of medication
|
|
0.743
|
|
No
|
1.133 (0.536–2.394)
|
|
|
Yes
|
1
|
|
|
Epworth Sleepiness Scale
|
|
0.043
|
|
Normal
|
0.527 (0.283–0.981)
|
|
|
Excessive Sleepiness
|
1
|
|
Notes: aCalculated with a 95% confidence interval; blogistic regression model, including all studied covariates; 1= ref.
Discussion
Sleep is fundamental for health, and it is closely interconnected with other diseases.
A sleep disorder compromises an individual's quality of life. The use of medications
to mitigate the consequences of sleep deprivation is increasingly common; however,
the impact caused is not always positive, since medication interferes directly (through
neurotransmitters with an impact on sleep) or indirectly on sleep, and it may potentiate
the opposite effect.[9] In addition, it is known that shift work has a negative impact on sleep quality
and, consequently, on quality of life.[8]
The aim of the present study was to evaluate the association between medication use
and sleep quality in shift workers versus daytime workers. Our results revealed that,
overall, there is no difference in sleep quality in those taking medication on a daily
basis compared to those who do so sporadically. In the study conducted by Kumar et
al.,[11] almost a third of the participants (who were taking medications) presented very
poor sleep quality, and the authors stated that the use of medications with sedative
and anticholinergic effects can contribute to an effect that is the opposite of its
purpose, since they can cause sedation and excessive sleepiness during the day. In
another study, Karami et al.[12] reported that the use of sedative medications improved sleep quality. In the present
study, when we take into consideration the specific medications that the workers were
taking, when such medications had a proven effect on sleep, in fact, sleep quality
was worse. Additionally, for those who already have taken sleeping medications, sleep
quality was worse. Although the directionality of this association is unclear (is
poor sleep quality the consequence of the use of such drugs or are those drugs being
used to improve poor sleep quality?), the fact remains that there was an association
found between the use of drugs with proven effects on sleep and poor sleep quality.
Fadhel[13] reported that individuals dependent on a medication presented more sleep problems,
making this a two-way relationship, as each problem can be the cause and consequence
of the other. Gordon[14] corroborates this idea, since in his study there was a higher prevalence of individuals
who still experienced sleep disturbances a long time after they had stopped using
medications with proven effects on sleep after other withdrawal symptoms had disappeared.
The use of substances such as stimulant drinks and stimulant medications have been
associated with worse sleep quality, which worsens the more often those substances
are consumed,[13] just like the results of the present study show. Some of these substances, such
as coffee, alcohol, energy drinks and medications are often used by shift workers
to improve the symptoms of sleepiness and poor sleep quality, as a consequence of
the alteration in the circadian rhythm.[15]
Male workers presented worse sleep quality, which can be partially justified by the
fact that, in the present study, men were older than women. Age has been described
as a risk factor for poor sleep quality, with the occurrence of sleep disorders increasing
with age, causing changes in sleep and affecting its quality.[10]
The preexistence of CV risk factors was also associated with poor sleep quality. This
is also a two-way association, since sleep disorders can increase the risk of CV events,[16]
[17] obesity, type-2 diabetes, and atherosclerosis.[18]
For those who were previously diagnosed with a psychological disorder, sleep quality
was worse, which is aligned with the findings of Slaven et al.,[19] Kumar et al.,[11] and Kalmbach et al.,[20] in which sleep quality was associated with depressive symptoms and even depression.
In fact, in the present study, all workers who had been diagnosed with a sleep disorder
had poor sleep quality.
The results of the present study support the published evidence of the association
between type of work and sleep quality, with shift workers presenting worse sleep
quality. Kerkhof[21] reported a higher prevalence of general sleep disturbances in shift workers compared
to day workers. Shift work, particularly night work, can have a negative impact on
health and well-being, increasing the risk of sleep disorders,[22]
[23] as well as that of various somatic and other psychological health conditions.[23] The impact of circadian rhythm disturbances (such as those caused by night work
or shift work) on sleep quality is greater than that of non-modifiable factors such
as age.[10] Shift time has also been described as an important risk factor for sleep disorders,
since sustained or prolonged exposure to risk factors, whether biological, behavioral,
individual or social, probably results in a higher risk of long-term adverse consequences
compared to brief or short-term exposure.[24] However, the results of the present study showed that those who were working on
the current shift for up to five years had worse sleep quality. This can imply some
sort of adjustment in the sleeping habits of these workers, in which their experience
helps them adjust their routines to reduce the impact that shift work can have on
sleep quality, just like it was found by Costa.[25]
After adjusting all variables associated with sleep quality in the bivariate analysis,
most variables did not increase the risk of having poor sleep quality, since the calculated
odds ratio was not statistically significant, and this is the main finding and strength
of the present study. The published literature on sleep quality mainly focuses on
the individual effect of a specific variable on sleep, usually by conducting a bivariate
analysis. But, since sleep quality is a multidimensional problem, influenced by multiple
factors (related to the individual, to their working conditions, to the use of different
substances, and others), it should be analyzed considering different factors, adjusted
to one another, just we like did in the present study.
Of course, these results must be weighed against some limitations. First of all, the
present was an observational study with no randomization of participants. Our results
cannot be generalized to every activity sector. Secondly, the sampling method we used
can lead to selection bias, since only those accessing their email or social media
accounts could access the link to participate in the study. Also, the sample size
may have influenced the results of our logistic model. Finally, in spite of the fact
that the present study included a multiplicity of factors that can influence sleep
quality, important factors, such as the time of exposure to light and/or the number
of successive shifts without rest, and the use of light to stay awake were not addressed,
since they were not collected for analysis.
Still, the current study presents some innovative features that should explored in
other professional settings, with larger sample sizes and, preferably, under controlled
conditions.
Conclusion
Sleep quality is such a complex issue that its analysis should be multifactorial,
not restricted to a simple association between a single variable and sleep quality.
Although there were several factors that individually negatively influenced sleep
quality, when adjusted to one another, by using a logistic regression model, they
did not increase the risk of having poor sleep quality. No differences were found
regarding medication use, type of work, and sleep quality.
Understanding and promoting sleep quality and its underlying factors is a key factor
to avoid pharmacological sleep iatrogenesis, encourage the rational and safe use of
medications, and thereby improve overall health. Further research is necessary to
confirm our findings, since they are restrained by the limitations pf the present
study.
