Keywords
inadequate sleep - armed forces - fatigue - operational capacity - military personnel
- national security
Introduction
Sleep is a cyclical behavior characterized by temporary and reversible changes in
mobility and especially in consciousness, being essential for maintaining homeostasis.[1] Alteration in their quantity or quality can impair an individual's occupational
performance and activities of daily living,[2] in addition to being associated with several health problems, including increased
risk of cardiovascular disease, diabetes, and obesity.[3]
Several professional categories are subject to inadequate sleep, including the military.
Factors such as increased physiological and psychological demands,[4] military rank and marital status,[5] care for young children[2] and the work routine itself[6] can pose challenges to getting enough restful sleep. This can culminate in a high
prevalence of poor sleep quality among these professionals.[7] This inadequacy can induce negative physiological and behavioral effects, increase
reaction time, and worsen performance in visuospatial tasks[8] increasing the chances of errors and accidents. It can also impact physical performance,
aggravate injuries, generate cognitive impairment, and interfere with leadership,
and the ability to make decisions and carry out planning. Together, they all negatively
influence the operational capacity of the military. However, modifiable factors such
as physical exercise,[9] adequate nutrition[10] and increased mental resilience[11] are associated and contribute to better sleep quality.
In this context, evaluating the quality of sleep of the Marine Corps (Corpo de Fuzileiros Navais - CFN) - the armed wing of the Brazilian Navy, a group dedicated to amphibious and expeditionary
operations - is necessary, since studies involving a profile on the quality of sleep
of its military personnel have not been conducted in its 216 years of existence.
Given the lack of work examining this issue in the CFN in Brazil, an exploratory study is needed to determine the true extent of the problem
to develop interventions (if necessary) to improve sleep based on evidence; these
interventions would then be aimed at maintaining the readiness of this military personnel.
Therefore, the objective of this work was to evaluate the sleep quality and associated
factors in the Marine Corps of the Brazilian Navy.
Materials and Methods
Participants
This is a cross-sectional study developed with CFN soldiers, subordinate to the Fleet Marine Force, on active duty, and serving in the city of Rio de Janeiro, between June and July
2022.
The OpenEpi online version 3.01 program was used to calculate the sample size. An
estimated population of 6,475 military personnel was considered, confidence level
of 95%, expected prevalence of 58.8%,[7] predicted sampling error of 3, design effect of 1, and supplementation of 20% for
cover losses and control for confounding factors, thus requiring a minimum sample
of 1,079 individuals.
Participation was voluntary and each soldier answered the questionnaires only after
reading and signing the informed consent form in electronic format. Incoherent responses
in any questionnaire were disregarded and only questionnaires that were fully answered
were counted. The study was approved by the Ethics Committee for Research with Human
Beings (CAAE: 53174321.7.0000.5256/Protocol Number: 5.202.697; Marcílio Dias Naval
Hospital/Rio de Janeiro).
Operational Procedures
The LimeSurvey® software program was used to administer the questionnaires, with operational procedures
guided by the CHERRIES checklist.[12] Its objective was explained, and doubts were previously resolved in all participating
Military Organizations before distributing the survey link. Cookies were used for
access control, allowing participants to save partially answered questionnaires to
be completed later. Manual verification of each response was performed to detect duplicates
and inconsistencies. A pilot study was performed aiming at adjustments and standardization.
Instruments
Quality of Sleep
Sleep quality, a dependent variable, was assessed using the Brazilian version of the
Pittsburgh Sleep Quality Index (PSQI).[13] This questionnaire contains 19 items that assess the subjective quality of sleep
over the last month from 7 domains. Its score ranges from 0 to 21, with values up
to 5 indicating good sleep quality and values above 5 indicating poor sleep quality.
Sociodemographic, Behavioral, and Professional Information
Sociodemographic information was collected (age; gender; education; marital status;
self-reported skin color; if the individual lived with children up to five years old;
behavioral information (use of tobacco and alcohol; time of day when one has more
energy; health restriction in the last month consumption of stimulating drinks in
the afternoon); and specific information about the military profession (rank; participation
in military exercises lasting more than 72 hours in the last month; problems at work
due to sleep whether they worked in operational or administrative work).
Sleepiness
Sleepiness was assessed using the Brazilian version of the Epworth Sleepiness Scale
(ESE).[14] There are eight questions that verify the probability of the respondent to doze
off in different daily situations, active or passive. Scores range from 0 to 3, with
0 - never dozing off; 1 - a low chance of dozing off; 2–a medium chance of dozing
off; and 3–a high chance of dozing off. Values lower than or equal to 10 in the overall
score indicate normal sleepiness, between 11 and 15 excessive daytime sleepiness (EDS),
and values greater than or equal to 16 as very EDS.
Physical Activity and Sedentary Behavior
Physical activity and sedentary behavior were verified using the International Physical
Activity Questionnaire (IPAQ; short version), validated in Brazil.[15] The physical activity level was estimated by the time spent in moderate activities
plus twice the time in vigorous physical activity. Participants who reported 150 minutes
or more of physical activity in the week were classified as physically active, and
those with less than 150 minutes were considered physically inactive. Sedentary behavior
was assessed by sitting time investigated by the question “How much time in a day
do you sit on a weekday/the weekend?” Sitting time on a weekday was multiplied by
five, and on weekends by two; the value found was later divided by seven to define
the daily average (minutes) spent sitting.
Eating Habits
Eating habits were evaluated using the scale by Gabe & Jaime (2019).[16] The instrument evaluates eating practices through four dimensions: planning, domestic
organization, food choice, and eating habits, represented by a set of 24 items that
exemplify such practices. Categorization was performed by analysis of tertiles, in
which scores greater than 41 indicate excellent eating habits; between 32 and 41 normal;
and less than 32 as insufficient.[17]
Distress
The Kessler Psychological Distress Scale (K10),[18] validated in Brazil, was used to assess chronic stress through 10 questions about
manifestations of distress in the last 30 days. The score ranges from 10 to 50, with
values between 0 and 15 meaning absent or low mental disorder; between 16 and 21,
moderate; between 22 and 29, severe; and between 30 and 50, very severe mental disorder.
Anthropometric Measurements
Secondary data belonging to the database of the Military Aspect Program of the Brazilian
Navy, which only had the values of male soldiers, were used. This program evaluates
the military aspect of FN using scientifically validated tests to measure individual body composition. Measurements
were obtained as described in the Norms on Military Physical Training and Physical
Assessment Tests in the Brazilian Navy.[19] All assessments were conducted in September and October 2022.
Measurements of height (m), total body mass (kg), waist circumference (cm)[20] and skinfolds[21] were considered. The Body Mass Index (BMI; Kg/m2) was estimated, and classified according to the World Health Organization[22]; the waist-to-height ratio (WHR), adopting a cut-off point of ≥ 0.5 as increased[23]; the sum of skinfolds (∑DC) and body fat percentage (%F) was calculated by the Siri
equation[24].
Statistical Analysis
Data normality was verified by the Shapiro-Wilk test and graphically by histograms.
Continuous variables were characterized by calculating the mean ( x̅ )and standard deviation (SD), or median (Md) and interquartile range (IQR), depending
on normality. Categorical variables were presented by absolute (n) and relative (%)
frequency. The chi-squared test and independent t-test or Mann-Whitney test were used to compare groups with good (score ≤ 5) and poor
(score >5) sleep quality, depending on data normality.
Poisson regression with robust variance was used to determine the factors associated
with sleep quality. The associated factors were first investigated in a crude model
(bivariate analysis), and subsequently in a model adjusted by covariates. The variables
were included in the multiple models by the stepwise backward method when they presented
p ≤ 0.20 in the bivariate model. After analyzing the independent variables in the
multiple regression model, the variables that were associated with the outcome with
p < 0.05 were maintained in the final model and considered independently associated.
Prevalence ratios (PR) and their respective 95%CI were calculated. All statistical
analyses were performed using the Stata statistical version 14.0 software program.
Results
The data analysis flowchart is shown in [Fig. 1]. Among the 1,553 research participants, 1,248 individuals correctly completed the
PSQI.
Fig. 1 Selection and analysis sample flowchart.
Almost all the sample was made up of men, which is why the data were analyzed in aggregate
considering both genders. The approximate age of the participants was 30 years old,
more than half consumed alcohol, most did not smoke, were physically active, had a
BMI classified as overweight and 925 of them (74.12%) had poor sleep quality.
Significant differences were observed in the comparison by subgroup between sleep
quality and age, living with children younger than five years, physical activity level,
daytime sleepiness, distress, eating habits, period of the day with greater energy,
routine, problems at work, rank and type of work ([Table 1]).
Table 1
Sociodemographic, anthropometric, behavioral, and work characteristics of Marines
|
Variable
|
Good sleep quality
(Score ≤ 5)
|
Poor sleep quality
(Score >5)
|
p-value
|
Total
|
|
Age (years)[a]
|
33.00 (27.00–41.00)
|
30.00 (26.00–37.00)
|
<0.001[a]
|
31.00 (26.00–38.00)
|
|
Gender[c]
|
|
|
|
|
|
Female
|
3 (0.93)
|
8 (0.86)
|
0.916[c]
|
11 (0.88)
|
|
Male
|
320 (99.07)
|
917 (99.14)
|
1.237 (99.12)
|
|
Education[c]
|
|
|
|
|
|
Completed high school
|
173 (53.56)
|
492 (53.19)
|
0.908 [c]
|
665 (53.29)
|
|
Incomplete/complete higher education
|
150 (46.44)
|
433 (46.81)
|
583 (46.71)
|
|
Civil status[c]
|
|
|
|
|
|
Single
|
94 (29.10)
|
319 (34.49)
|
0.077 [c]
|
413 (33.09)
|
|
With companion
|
229 (70.90)
|
606 (65.51)
|
835 (66.91)
|
|
Skin color[c]
|
|
|
|
|
|
White
|
105 (32.51)
|
258 (27.89)
|
0.239 [c]
|
363 (29.09)
|
|
Black
|
43 (13.31)
|
145 (15.68)
|
188 (15.06)
|
|
Other
|
175 (54.18)
|
522 (56.43)
|
697 (55.85)
|
|
Smoke[c]
|
|
|
|
|
|
No
|
296 (91.64)
|
834 (90.16)
|
0.434 [c]
|
1.130 (90.54)
|
|
Yes
|
27 (8.36)
|
91 (9.84)
|
118 (09.46)
|
|
Consume alcohol[c]
|
|
|
|
|
|
No
|
144 (44.58)
|
362 (39.14)
|
0.076 [c]
|
506 (40.54)
|
|
Yes
|
179 (55.42)
|
563 (60.86)
|
742 (59.46)
|
|
Live with child(ren) ≤ 5 years of age[c]
|
|
|
|
|
|
No
|
246 (76.16)
|
625 (67.57)
|
0.004 [c]
|
871 (69.79)
|
|
Yes
|
77 (23.84)
|
300 (32.43)
|
377 (30.21)
|
|
Weight (kg)[b]
|
82.44 (11.95)
|
83.63 (13.06)
|
0.159 [b]
|
83.32 (12.79)
|
|
Height (m)[b]
|
1.76 (0.07)
|
1.76 (0.06)
|
0.759 [b]
|
1.76 (0.06)
|
|
BMI (kg/m2)[b]
|
26.67 (3.45)
|
27.08 (3.76)
|
0.089 [b]
|
26.97 (3.69)
|
|
BMI[c]
|
|
|
|
|
|
Underweight or eutrophic
|
101 (32.37)
|
259 (29.27)
|
0.335 [c]
|
360 (30.08)
|
|
Overweight
|
161 (51.60)
|
454 (51.29)
|
615 (51.37)
|
|
Obese
|
50 (16.03)
|
172 (19.44)
|
222 (18.55)
|
|
Waist circumference (cm)[b]
|
87.16 (8.76)
|
87.72 (9.16)
|
0.346 [b]
|
87.57 (9.06)
|
|
Waist circumference[c]
|
|
|
|
|
|
Normal
|
199 (63.78)
|
534 (60.34)
|
0.277 [c]
|
733 (61.24)
|
|
Increased
|
113 (36.22)
|
351 (39.66)
|
464 (38.76)
|
|
Sum of folds (mm)[b]
|
53.38 (21.98)
|
53.60 (23.13)
|
0.881 [b]
|
53.54 (22.83)
|
|
Fat percentage[b]
|
16.32 (6.09)
|
16.21 (6.50)
|
0.798 [b]
|
16.24 (6.39)
|
|
Waist-to-height ratio[b]
|
0.50 (0.05)
|
0.50 (0.05)
|
0.281 [b]
|
0.50 (0.05)
|
|
Waist-to-height ratio[c]
|
|
|
|
|
|
Normal
|
167 (53.53)
|
443 (50.06)
|
0.292 [c]
|
610 (50.96)
|
|
Increased
|
145 (46.47)
|
442 (49.94)
|
587 (49.04)
|
|
NAFTS - (min)[a]
|
474.00 (230.00–790.00)
|
460.00 (220.00–770.00)
|
0.653[a]
|
460.00 (220.00–780.00)
|
|
NAFTS[c]
|
|
|
|
|
|
Physically inactive
|
39 (12.70)
|
159 (17.87)
|
0.036 [c]
|
198 (16.54)
|
|
Physically active
|
268 (87.30)
|
731 (82.13)
|
999 (83.46)
|
|
Time seated (min)[a]
|
206.00 (123.00–329.00)
|
197.00 (120.00–300.00)
|
0.282[a]
|
197.00 (120.00–309.00)
|
|
Sleepiness level[c]
|
|
|
|
|
|
Normal
|
266 (84.18)
|
522 (57.49)
|
<0.001 [c]
|
788 (64.38)
|
|
EDS
|
39 (12.34)
|
253 (27.86)
|
292 (23.86)
|
|
Very EDS
|
11 (3.48)
|
133 (14.65)
|
144 (11.76)
|
|
Distress[c]
|
|
|
|
|
|
Absent or low mental disorder
|
258 (83.23)
|
380 (43.08)
|
<0.001[c]
|
638 (53.52)
|
|
Moderate mental disorder
|
33 (10.64)
|
264 (29.93)
|
297 (24.92)
|
|
Severe mental disorder
|
16 (5.16)
|
153 (17.35)
|
169 (14.18)
|
|
Very severe mental disorder
|
3 (0.97)
|
85 (9.64)
|
88 (7.38)
|
|
Eating habits[c]
|
|
|
|
|
|
Excellent
|
129 (41.61)
|
247 (27.85)
|
<0.001
|
376 (31.41)
|
|
Normal
|
140 (45.16)
|
479 (54.00)
|
619 (51.71)
|
|
Insufficient
|
41 (13.23)
|
161 (18.15)
|
202 (16.88)
|
|
Stimulating drinks in the afternoon[c]
|
|
|
|
|
|
No
|
117 (36.22)
|
302 (32.65)
|
0.242[c]
|
419 (33.57)
|
|
Yes
|
206 (63.78)
|
623 (67.35)
|
829 (66.43)
|
|
Period with more energy[c]
|
|
|
|
|
|
Morning
|
256 (79.26)
|
610 (65.95)
|
<0.001[c]
|
866 (69.39)
|
|
Afternoon
|
49 (15.17)
|
205 (22.16)
|
254 (20.35)
|
|
Night
|
18 (5.57)
|
110 (11.89)
|
128 (10.26)
|
|
Normal routine[c]
|
|
|
|
|
|
No
|
19 (5.88)
|
88 (9.51)
|
<0.001[c]
|
107 (8.57)
|
|
Yes
|
304 (94.12)
|
837 (90.49)
|
1.141 (91.43)
|
|
Problems at work[c]
|
|
|
|
|
|
No
|
272 (84.21)
|
430 (46.49)
|
<0.001[c]
|
702 (56.25)
|
|
Yes
|
51 (15.79)
|
495 (53.51)
|
546 (43.75)
|
|
Rank[c]
|
|
|
|
|
|
SD/Pr
|
108 (33.44)
|
443 (47.89)
|
<0.001[c]
|
551 (44.15)
|
|
SG/SO
|
178 (55.10)
|
432 (46.70)
|
610 (48.88)
|
|
Official
|
37 (11.46)
|
50 (5.41)
|
87 (6.97)
|
|
Type of work[c]
|
|
|
|
|
|
Administrative
|
147 (45.51)
|
355 (38.38)
|
<0.001[c]
|
502 (40.22)
|
|
Operative
|
176 (54.49)
|
570 (61.62)
|
746 (59.78)
|
|
Handling > 72h in the last month[c]
|
|
|
|
|
|
No
|
203 (62.85)
|
556 (60.11)
|
0.242[c]
|
2759 (60.82)
|
|
Yes
|
120 (37.15)
|
369 (39.89)
|
489 (39.18)
|
Abbreviations: BMI, Body Mass Index; EDS, excessive daytime sleepiness; NAFTS, Total
Weekly Physical Activity Level; SD/Pr, soldier/private; SG/SO, sergeant/non-commissioned
officer.
a = quantitative variables with non-normal distribution [Median (Interquartile range),
Mann-Whitney test].
b = quantitative variables with normal distribution [Mean ± (Standard deviation),
Independent t-test].
c = categorical variables [n (%), Chi-Squared Test]; BMI: underweight or eutrophic
≤ 24.9 kg/m2; overweight = 25 kg/m2–29.9 kg/m2; obese ≥ 30 kg/m2; Waist circumference:
men: ≥ 90 cm = increased; women: ≥ 80 cm = increased; Waist-to-height ratio: ≥ 0.5 = increased;
bold = p < 0.05.
[Table 2] presents the bivariate analysis related to poor sleep quality, while [Table 3] shows the associated covariates in the adjusted model. It was identified that age,
being an officer and being physically active are protective factors (negative association)
of poor sleep quality. On the other hand, living with children younger than 5 years
old, having more energy at night, believing that they have problems at work due to
sleep, being excessively sleepy, having a mental disorder, and having normal eating
habits are predisposing factors (positive association) to poor sleep quality ([Table 3]).
Table 2
Factors associated with poor sleep quality in Marines – crude model – (bivariate analysis)
|
Variable (n = 1,248)
|
PR
|
95%CI
|
p-value
|
|
Age
|
0.99
|
0.98–0.99
|
< 0.001
|
|
Education
|
|
|
|
|
Completed high school
|
1.00
|
−
|
−
|
|
Completed/incomplete higher education
|
1.00
|
0.94–1.07
|
0.908
|
|
Civil status
|
|
|
|
|
Single
|
1.00
|
−
|
−
|
|
With companion
|
0.94
|
0.88–1.00
|
0.068
|
|
Smoke
|
|
|
|
|
No
|
1.00
|
−
|
−
|
|
Yes
|
1.04
|
0.94–1.16
|
0.409
|
|
Consume alcohol
|
|
|
|
|
No
|
1.00
|
−
|
−
|
|
Yes
|
1.06
|
0.99–1.14
|
0.092
|
|
Live with children under 5 years of age
|
|
|
|
|
No
|
1.00
|
−
|
−
|
|
Yes
|
1.11
|
1.04–1.18
|
0.002
|
|
Rank
|
|
|
|
|
SD/Pr
|
1.00
|
|
−
|
|
SG/SO
|
0.88
|
0.82–0.94
|
< 0.001
|
|
Officers
|
0.71
|
0.59–0.86
|
< 0.001
|
|
Type of work
|
|
|
|
|
Administrative
|
1.00
|
|
−
|
|
Operative
|
1.08
|
1.01–1.16
|
0.028
|
|
Period with more energy
|
|
|
|
|
Morning
|
1.00
|
−
|
−
|
|
Afternoon
|
1.15
|
1.06–1.24
|
< 0.001
|
|
Night
|
1.22
|
1.12–1.32
|
< 0.001
|
|
Problems at work
|
|
|
|
|
No
|
1.00
|
−
|
−
|
|
Yes
|
1.48
|
1.39–1.58
|
< 0.001
|
|
Normal routine
|
|
|
|
|
No
|
1.00
|
−
|
−
|
|
Yes
|
0.89
|
0.81–0.98
|
0.018
|
|
Handling >72h
|
|
|
|
|
No
|
1.00
|
−
|
−
|
|
Yes
|
1.03
|
0.96–1.10
|
0.381
|
|
Level of sleepiness (
n
= 1,224)
|
|
|
|
|
Normal
|
1.00
|
−
|
−
|
|
EDS
|
1.31
|
1.22–1.40
|
< 0.001
|
|
Very EDS
|
1.39
|
1.30–1.49
|
< 0.001
|
|
NAFTS (
n
= 1,197)
|
1.00
|
1.00–1.00
|
0.700
|
|
NAFTS (
n
= 1,197)
|
|
|
|
|
Physically inactive
|
1.00
|
−
|
−
|
|
Physically active
|
0.91
|
0.84–0.99
|
0.020
|
|
Time seated (
n
= 942)
|
1.00
|
1.00–1.00
|
0.300
|
|
Distress (
n
= 1,192)
|
|
|
|
|
Absent or low mental disorder
|
1.00
|
−
|
−
|
|
Moderate mental disorder
|
1.49
|
1.38–1.61
|
< 0.001
|
|
Severe mental disorder
|
1.52
|
1.40–1.65
|
< 0.001
|
|
Very severe mental disorder
|
1.62
|
1.50–1.75
|
< 0.001
|
|
Eating habits (
n
= 1,197)
|
|
|
|
|
Excellent
|
1.00
|
−
|
−
|
|
Normal
|
1.15
|
1.06–1.25
|
< 0.001
|
|
Insufficient
|
1.20
|
1.10–1.32
|
< 0.001
|
|
Stimulating drinks in the afternoon
|
|
|
|
|
No
|
1.00
|
−
|
−
|
|
Yes
|
1.04
|
0.97–1.12
|
0.251
|
|
BMI (
n
= 1,197)
|
1.09
|
0.99–1.03
|
0.385
|
|
BMI category (
n
= 1,197)
|
|
|
|
|
Underweight or Eutrophic
|
1.00
|
−
|
−
|
|
Overweight
|
1.03
|
0.95–1.11
|
0.528
|
|
Obese
|
1.08
|
0.98–1.19
|
0.130
|
|
Waist circumference (
n
= 1,197)
|
1.00
|
1.00–1.01
|
0.335
|
|
Fat percentage (
n
= 1,197)
|
1.00
|
0.99–1.00
|
0.792
|
|
Waist-to-height ratio (
n
= 1,197)
|
1.42
|
0.76–2.68
|
0.273
|
Abbreviations: BMI, Body Mass Index; EDS, excessive daytime sleepiness; NAFTS, Total
Weekly Physical Activity Level; SD/Pr, soldier/private; SG/SO, sergeant/non-commissioned
officer; bold, p < 0.2.
Table 3
Factors associated with poor sleep quality in Marines – adjusted model – (multivariate
analysis)
|
Variable (n = 1,248)
|
PR
|
95%CI
|
p-value
|
|
Age (years)
|
0.99
|
(0.98–0.99)
|
0.016
|
|
Live with children ≤ 5 years of age
|
|
|
|
|
No
|
1.00
|
−
|
−
|
|
Yes
|
1.10
|
1.03–1.17
|
0.002
|
|
Rank
|
|
|
|
|
SD/Pr
|
1.00
|
−
|
−
|
|
SG/SO
|
1.09
|
0.98–1.20
|
0.101
|
|
Officers
|
0.81
|
0.67–0.98
|
0.032
|
|
Period with more energy
|
|
|
|
|
Morning
|
1.00
|
−
|
−
|
|
Afternoon
|
1.06
|
0.99–1.14
|
0.114
|
|
Night
|
1.12
|
1.04–1.21
|
0.003
|
|
Problems at work
|
|
|
|
|
No
|
1.00
|
−
|
−
|
|
Yes
|
1.26
|
1.19–1.35
|
< 0.001
|
|
Sleepiness level (
n
= 1,224)
|
|
|
|
|
Normal
|
1.00
|
−
|
−
|
|
EDS
|
1.14
|
1.06–1.21
|
< 0.001
|
|
Very EDS
|
1.11
|
1.04–1.19
|
0.002
|
|
NAFTS (
n
= 1,197)
|
|
|
|
|
Physically inactive
|
1.00
|
−
|
−
|
|
Physically active
|
0.92
|
0.85–0.99
|
0.036
|
|
Distress (
n
= 1,192)
|
|
|
|
|
Absent or low mental disorder
|
1.00
|
−
|
−
|
|
Moderate mental disorder
|
1.36
|
1.27–1.47
|
< 0.001
|
|
Severe mental disorder
|
1.26
|
1.16–1.37
|
< 0.001
|
|
Very severe mental disorder
|
1.34
|
1.23–1.45
|
< 0.001
|
|
Eating habits (
n
= 1,197)
|
|
|
|
|
Excellent
|
1.00
|
−
|
−
|
|
Normal
|
1.09
|
1.01–1.18
|
0.035
|
|
Insufficient
|
1.00
|
0.91–1.10
|
0.986
|
Abbreviations: CI, confidence interval; EDS, excessive daytime sleepiness; NAFTS,
Total Weekly Physical Activity Level; PR, prevalence ratio; SD/Pr, soldier/private;
SG/SO, sergeant/non-commissioned officer.
Results only refer to individuals with complete data on all instruments used. bold = p < 0.05.
Discussion
This study aimed to evaluate the quality of sleep and its associated factors in Brazilian
FN (Marines). As the main results, we found a high prevalence (74.12%) of military personnel
with poor sleep quality and the existence of significant differences between sleep
quality and sociodemographic, behavioral, and professional factors ([Table 1]). Protective factors for sleep quality were age, being an officer, and being physically
active. Living with children younger than five years old, having more energy at night,
thinking they had problems at work due to sleep, EDS, mental disorders, and having
normal eating habits were predisposing factors to poor sleep quality ([Table 3]). No associations were observed between poor sleep quality and anthropometric measurements.
Quality of Sleep Analysis
Sleep quality is a subjective and multifactorial concept that considers quantitative
and qualitative aspects,[25] with the PSQI being the recurrent instrument for assessing this variable in different
populations. Several studies have reported different prevalences of poor sleep quality
in military personnel, such as Wang et al. (2020)[26] reporting 40.9% poor sleep quality in the Chinese military; Roustaei et al. (2017)[27] found 48% in the Iranian military; and Plumb et al. (2014)[28] 89.1% in US military participants in conflicts. Publications on sleep parameters
in Brazilian military personnel, in addition to being scarce, involve a reduced sample
size and are not uniform in terms of the method used, making comparisons difficult.
However, our results were similar to those of Iahnke et al. (2022),[29] Oliveira (2020)[30] and Pinto et al. (2018),[31] who found 66.2%, 81.59% and 63.6% prevalence of poor sleep quality using the PSQI
in army personnel, firefighters and military police, respectively. The fact that our
study took place shortly after the COVID-19 pandemic may have contributed to the high
prevalence of poor sleep, showing a final reflection of social changes such as confinement
and social isolation, less exposure to sunlight, reduced activity physical activity,
and increased stress.[32] Lack of knowledge about the importance and need for care for sleep hygiene by the
Brazilian military may have also contributed to the result found.
Subgroup Analysis
There were significant differences in the subgroup analysis. Sociodemographic,[33] behavioral[8] and professional[6] aspects were related to sleep quality. Among the tested variables, only having a
normal routine and the type of work (administrative and operative) did not remain
significant in the regression model. However, in researching US Air Force soldiers,
Tvaryanas et al. (2018)[34] reported that insufficient sleep was related to limitations at work and lack of
physical tests, an aspect which can interfere with the routine in the barracks. Roustaei
et al. (2017)[27] reported that 62.9% of Iranian police officers in the operational sector had poor
sleep quality compared with 20.0% in the administrative sector. The demands related
to constant preparation, the risk of death inherent to operational activities, even
in training, and the physical overload can mean stressful and difficult factors for
good sleep quality in military personnel in the operational sector.
Protective Factors
Older marines had better sleep quality than younger ones. Our study shows that the
prevalence of poor sleep quality decreased by 1% for each year of increasing military
age (PR 0.99 [95%CI: 0.98–0.99]). The study by Choi et al. (2022)[7] found that advancing age is a risk factor for poor sleep quality (OR 1.11 [95%CI:
1.05–1.17]), increasing the chance of this occurrence by 11%. The studies by Chou
et al. (2016)[35] and Plumb et al. (2014)[28] found no association between age and sleep quality in the regression analysis. These
results show a lack of consensus regarding the association of age with sleep quality,
requiring further research. In the military context, it is relevant to consider that
the type of service of an individual changes with advancing age, which may include
a decrease in shift work, which makes it difficult to sleep.[36] Despite the statistical significance in our regression analysis, the small difference
in the medians in the subgroup analysis ([Table 1]) points to a possible clinical irrelevance of the outcome when considering sleep
ontogeny.
Being an officer was a protective factor (PR 0.81 [95%CI: 0.67–0.98]) for sleep quality.
A study by Caldwell et al.[5] (2019) pointed out that sleep disorders may be related to the ranks of officers,
especially older ones. Other studies in the literature did not contemplate the association
of sleep quality in officers.[7]
[35] Compared with soldiers, officers do not perform the same services in shifts, in
addition to having a higher education level and salary, which can improve sleep in
these individuals.[37] Furthermore, our study includes officers in a single category, not differentiating
according to their rank due to statistical representativeness; As a result, even though
they are officers, there are a large number of military personnel with low ranks in
the officer corps who do not have a command function; this can alter health aspects
such as less stress, less responsibility at work and even family issues, such as not
having children.
Regular practice of physical exercise is beneficial for sleep quality, a fact which
was also observed in our study (PR 0.92 [95%CI: 0.85–0.99]) and is, therefore, a non-pharmacological
strategy for treating sleep-related problems. There is evidence that moderate to vigorous
physical exercise can decrease sleep fragmentation and improve sleep latency,[9] contributing to the treatment and prevention of sleep-wake cycle disorders.
Risk Factors
Factors unrelated to military life can influence combatants' sleep. Parents of children
younger than six years of age may suffer from sleep fragmentation. Mysliwiec et al.
(2021)[2] reported that the birth or adoption of a child is one of the factors that most negatively
affect the sleep of American military personnel. According to our research, military
personnel living with children up to five years old increased the prevalence of poor
sleep quality by 10% (PR 1.10 [95%CI: 1.03–1.17]). The challenges inherent to night
rest for parents of young children can be exacerbated by the existing difficulties
for military personnel to sleep. Furthermore, considering the high prevalence of poor
sleep quality in our sample, an increase in their stress level may occur, which also
influences the child and may start a vicious circle of poor sleep quality at home.[38]
The prevalence of individuals with poor sleep quality was significant among those
who had more energy at night (PR 1.12 [95%CI: 1.04–1.21]). The need to sleep is mediated
by external (light-dark cycle) and internal (physiological processes) factors.[39] Circadian preference was verified using a direct question about the period in which
the soldier thought they had more energy. Individuals with nocturnal preferences feel
better at night and sleep later.[40] Although we did not investigate other social issues such as watching television,
using cell phones, following sports games, and playing online games, these are increasingly
prevalent today and are factors of satisfaction. Case studies conducted with American
Marines noted that excessive use of video games (30 hours or more per week, beyond
a 40-hour or more workday) may be associated with sleep deprivation, resulting in
poor job performance and mood disorders.[41] Work in the Brazilian Navy starts at 07:30 am in the Marine Force Squad, and many military personnel have to wake up even earlier
due to commuting to the barracks. This may be related to poor sleep quality in these
military personnel and cause bias when they must do daytime tasks.[42] In addition, a study by Tonon et al. (2019)[43] in Brazilian military states that afternoon individuals were those who had more
depressive symptoms (PR 2.58 [95%CI: 1.54–4.33]), which may contribute to poor sleep
quality. However, the preference for the nocturnal period would not be a risk factor,
but the chronodisruption it causes would.[43]
Judging that you had problems at work due to sleep increased the prevalence of poor
sleep quality in the sample (PR 1.26 [95%CI: 1.19–1.35]). This is a result that reveals
the subjective perception of the influence of sleep on the military. Negative self-perception
of sleep quality is associated with difficulty concentrating (OR 1.73 [95%CI: 1.13–2.65]),
regardless of the number of hours of continuous sleep. Issues such as emotional exhaustion,[44] pain, hypervigilance, awakenings, nightmares, and even the fear of sleeping to avoid
bad dreams[28] may be related to subjective complaints about sleep. Comparing the prevalence found
of poor sleep quality (74.12%) and its relationship with the prevalence of individuals
who believe they have problems at work due to sleep (26% - [Table 3]), a lack of knowledge about sleep characteristics and its importance for well-being
and operability are perceived, as many military personnel may have an impact on this
poor quality of sleep and do not attribute it to it.
Excessive sleepiness occurs from increased pressure to sleep due to circadian (decreased
core temperature) or homeostatic (sleep deprivation) issues.[39] EDS is the propensity to sleep under circumstances that would be inappropriate for
the affected individual. It relates to lower cognition, which influences decision-making,
reaction time, and information processing.[45] Both EDS (PR 1.14 [95%CI: 1.06–1.21]) and severe EDS (PR 1.11 [95%CI: 1.04–1.19])
were positively associated with low sleep quality. The demands related to constant
preparation can mean stressful factors that hinder good quality sleep.[4] In addition, poor sleep quality may be a risk factor for sleepiness[46] and fatigue,[47] which may contribute to a model that cycles back.
Studying the relationship between mental health and sleep in military personnel is
relevant as they face several stressors[11] such as exposure to combat, shift work, and changes in the type of work.[48] A study performed with the US military reported that troops who had higher levels
of mental problems were also those who had lower sleep quality.[49] Our work found that poor sleep quality was at least 26% more prevalent ([Table 3]) in individuals with any level of mental disorder in the K10 assessment. In addition,
Choi et al. (2022)[7] found a greater chance of poor sleep quality in military personnel at four stress
levels (low to very high), while a study by Kim et al. (2016)[50] in Korean military personnel showed an association between distress (K10) and moderate
(β = 2.789, p < 0.0001) and severe (β = 5.245, p < 0.0001) difficulty sleeping.
Regarding eating habits, the literature shows a promising association between healthy
eating habits and improvement in sleep quality and components,[10] although the molecular mechanisms of this relationship need to be elucidated. Our
study unexpectedly found regular eating habits to be a predisposing factor to poor
sleep quality. Disorders caused by changes in the body's circadian cycle influence
food consumption from changes in appetite and satiety, favoring weight gain.[51] Eating habits can be altered due to a greater window of opportunity to eat when
there is sleep loss.[52] This situation allows neuroendocrine and metabolic alterations, such as a reduction
in leptin levels and an increase in ghrelin levels.[51] This scenario may apply to our sample, since 51.29% of the subjects who had poor
sleep quality were also overweight ([Table 1]). However, a systematic review of the relationship between diet and sleep quality
concluded that the overall quality of the studies performed so far is poor to regular,
not allowing us to clearly define cause and effect relationships.[3] Thus, one of the reasons for our result may be due to the collection instrument
itself. The Eating Habits Scale used has a discrete numerical result. Categorization
was performed based on possible tertiles. Since this classification was not validated
together with the questionnaire, the statistical evaluation of this variable categorically
may have interfered with the relationship with sleep quality. As we performed a cross-sectional
study, the instrument only allows a subjective analysis of the availability of food
for consumption, in addition to not including questions about changes in appetite
and regular portion sizes, which are important aspects since increased appetite can
lead to increased caloric intake without affecting eating behaviors. Furthermore,
questionnaires often fail to predict nutritional status, presumably because individuals,
especially obese individuals, tend to underestimate their food intake.[53]
There was no association between anthropometric variables (waist circumference, fat
percentage, and waist-to-height ratio) in the bivariate analysis, and in particular
BMI in the multivariate analysis. The association between sleep quality and anthropometric
measures was non-significant. Although most of the sample was classified as overweight,
most military personnel had normal values for waist circumference and waist-to-height
ratio ([Table 1]). Additionally, a significant portion of the sample (83.46%) was physically active,
and the body fat percentage (16.24 ± 6.39%) ([Table 1]) was considered slightly better than the average according to the body composition
standards of the Brazilian Navy.[19] It is likely that Brazilian FN is more homogeneous in terms of anthropometric characteristics,
corroborating the physical fitness required for the success of military operations.[54] Therefore, Brazilian FN have sleep profiles that do not seem to be distinguished
according to body composition. In addition, as they are soldiers who make up the Navy's
special troops, they must present a certain physical activity level, anthropometric
standards, and a body composition profile specifically oriented toward the mission,
unlike other military units of the corps, or even soldiers from other forces.[55]
Our results also support the study by Rush et al. (2016)[56] showing that obesity is lower for Marines, and de Barlas et al. (2013)[57] who showed a lower obesity prevalence and higher physical activity rates in these
military personnel compared with other operational services of other Forces, indicating
that there are probably differences related to the activities performed. Another factor
that may have influenced this non-significant result is that we used secondary data
from a pre-existing databank in the electronic system of the Brazilian CFN herein. However, when extracting and analyzing the information, we considered the
normative documents that detail the methodological aspects of obtaining body measurements
and the possible implausible measures, ensuring greater reliability of the data available.
Limitations and Strengths
Limitations and Strengths
Our work had some limitations. Anthropometric variables were obtained using secondary
data from CFN systems in Brazil. The size of the female sample did not allow a stratified analysis
by gender. Due to the nature of a cross-sectional study, it was not possible to establish
a causal relationship between sleep quality and the investigated factors. The number
of responses from officials was small, which may interfere with the interpretation
of the data. In addition, objective measurements were not used in the study.
As strengths, we used validated questionnaires and scales that minimize biases and
enable reliable results, and their remote application, which allowed the analysis
of a large sample. Furthermore, as far as we know, this was the first study on sleep
quality in Brazilian FN, thereby to date constituting the largest epidemiological study on sleep in military
personnel ever performed in Brazil.
Conclusion
It is concluded that the Brazilian Navy Marines had a high prevalence of poor sleep
quality associated with personal, family, and occupational factors as contributors
to the problem, indicating the need to develop health actions that favor good sleep
hygiene in these professionals.
