Sleep Epidemiology 3 (2023) 100062

Contents lists available at ScienceDirect

Sleep Epidemiology
journal homepage: www.elsevier.com/locate/sleepe

A cross-sectional study of the association between sleep quality and anxiety

in postsecondary students in Ontario
Magdalena Albrecht-Bisset a, b, Dan Wang a, b, Krystle Martin a, c, Pierre Côté a, b, d,
Efrosini A. Papaconstantinou a, b, *
a
Faculty of Health Sciences, University of Ontario Institute of Technology (Ontario Tech University), 2000 Simcoe Street North, Oshawa ON, L1H 7K4, Canada
b
Institute of Disability and Rehabilitation Research, Oshawa, Ontario, Canada
c
Ontario Shores Centre for Mental Health Sciences, Whitby, Ontario, Canada
d
Canada Research Chair in Disability Prevention and Rehabilitation, Toronto, Canada

A R T I C L E I N F O A B S T R A C T

Keywords: Postsecondary students frequently report high rates of anxiety and poor sleep quality. The association between
Sleep quality poor sleep quality and anxiety is poorly understood in this population. We conducted a cross-sectional study of
Anxiety students enrolled in two faculties at Ontario Tech University (OTU) - Faculty of Health Sciences (FHS) and
Emerging adults
Faculty of Education (FEd), and students attending the Canadian Memorial Chiropractic College (CMCC) during
Postsecondary
Cross-sectional
the fall of 2017 to determine the prevalence of poor sleep quality and moderate to extremely severe anxiety, as
well as the association between them. Participants completed self-report questionnaires to measure sleep quality
(Pittsburgh Sleep Quality Index [PSQI]), anxiety (Depression, Anxiety, and Stress Scale – 21 Items [DASS-21]),
socio-demographic, lifestyle and health-related variables. Multivariable logistic regression was used to measure
the association between poor sleep quality and moderate to extremely severe anxiety, as well as to control for
covariates. A significant correlation between PSQI scores and DASS-21 anxiety scores was found in both pop­
ulations: students who reported poor sleep quality were more likely to report moderate to extremely severe
anxiety. This association decreased but still remained high even after controlling for important covariates (i.e.,
biological sex, stress, and depression).

Introduction students face when transitioning from secondary to postsecondary ed­

ucation including increased academic pressures, increased expectations
In Canada, rates of mental health issues among postsecondary stu­ and responsibilities and increased financial burden [3,12,13]. When
dents are significantly higher than among the general population with addressing mental health issues, such as anxiety, among students, it is
anxiety-related disorders being the most commonly reported [1–4]. This important to understand what (if any) mechanisms are contributing to
is causing wait times for both on-campus and community mental health it. One such mechanism that may play a role in anxiety (and other
services as long as two to three years at some institutions [1,3,5]. This is mental health issues) in postsecondary students is poor sleep quality.
not only a Canadian problem but rather the increasing rates of mental Sleep problems impact both the physical and mental health of
health issues among postsecondary students has become a public health postsecondary students and their daytime functioning. Postsecondary
issue around the world [4,6-11]. students regularly report high rates of poor sleep quality and in 2019,
In Canada, the one-year overall prevalence of all mental health issues the Canadian Reference Group (CRG) of the National College Health
among the general population is approximately 20% [2]. Specifically in Assessment (NCHA) found that 65% reported feeling tired or dragged
postsecondary students, in 2019 68% of postsecondary students re­ out the majority of days and 48% reported that daytime sleepiness
ported experiencing overwhelming anxiety in the past year, 51% re­ impacted them during activities [4]. Recent sleep research has put a
ported feeling so depressed it was difficult to function and 63% felt that specific focus on postsecondary students as a group that is separate from
things were hopeless [4]. The high rates of anxiety and other mental both children and adults [4,10,12]. Postsecondary students face a
health issues are thought to be related to the major life changes that unique combination of academic, family, work and social pressures

* Corresponding author.
E-mail address: efrosini.papaconstantinou@ontariotechu.ca (E.A. Papaconstantinou).

https://doi.org/10.1016/j.sleepe.2023.100062

Available online 20 July 2023

2667-3436/© 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
M. Albrecht-Bisset et al. Sleep Epidemiology 3 (2023) 100062

making them especially vulnerable for the development of sleep disor­ Sciences, Health Science, Kinesiology, Medical Laboratory Sciences and
ders, poor sleep habits and poor sleep quality [4,6,12,14]. Nursing Science. The FEd offers two main programs: Education
There is considerable change to students’ sleep patterns when they (including Primary/Junior, Junior/Intermediate and Intermediate/Se­
enter postsecondary education, such as delays in their sleep time, which nior Education) and Arts in Educations Studies and Digital Technology.
contributes to irregular sleep patterns. Changes to sleep patterns may be CMCC is a private postsecondary institution located in Toronto, Ontario.
attributed to many factors including early class start times, part-time It offers a second-entry four-year professional undergraduate profes­
jobs, lifestyle preferences, social activities and academic demands [6, sional degree (for students who have completed at least three years of
15]. The combination of these factors may impact the amount of time university study), as well as postgraduate and continuing education
that students have to sleep but this is often exacerbated in students who programs.
are juggling other responsibilities outside of school including caring for In 2017, 1931 students were enrolled in the FHS and 268 students in
family members, working part-time jobs and commuting [6]. Further­ the FEd, collectively making up 23% of the OTU undergraduate popu­
more, students may have misconceptions about the importance of sleep lation. Students enrolled in both the FHS and the FEd were primarily
and believe that the high levels of stress or anxiety and sleep deprivation domestic (98.3%, 99.6%) and female (75.9%, 66.7%), with a mean age
is an inevitable part of “normal” university life [10,16-19]. of 24 and 28 respectively. In the fall of 2017 there were 766 students
Understanding the association between poor sleep quality and anx­ enrolled in the undergraduate chiropractic program at CMCC. Students
iety is important since sleep problems can heighten anxiety and anxiety enrolled in this program had a mean age of 25 and with a close to even
can amplify sleep problems – which can create a vicious cycle [19–21]. distribution of males and females (55.4% female).
Anxiety activates the sympathetic nervous system, which leads to a
number of physiological responses including increased heart rate and Recruitment
breathing [22–24]. These physiological responses create a sense of
arousal which is sometimes referred to as hypervigilance [22–25]. These OTU
physiological changes, coupled with hypervigilance are counterpro­ The study was advertised to the FHS and FEd student population
ductive for sleep. Sleep onset can only occur if an individual is relaxed; using social media platforms (Facebook and Twitter). Flyers were also
without calming the sympathetic nervous system and increased activa­ posted around campus containing information about the study. Partic­
tion of the parasympathetic nervous system, individuals may have ipants were recruited in September and October of 2017. Recruitment
delayed sleep onset [26–28]. This can affect an individual’s sleep quality took place in three waves. In wave one, students were recruited through
and quantity, and impede daytime function [28,29]. Poor sleep quality an in-class recruitment strategy. Members of the research team attended
and quantity affect mood and make one more anxious, which in turn, classes in the FHS and FEd to formally inform students about the study
make it even more difficult to sleep the following sleep period [21, and invite them to participate. Following the presentation, students
30-32]. This creates a cycle that not only impacts an individual’s sleep were provided with a link to the online consent form and study ques­
quality and mental health but also can lead to negative academic, social tionnaire. A pre-recorded instructional video was distributed to online
and physical outcomes [22-24,30]. classes within these faculties.
A recent systematic review of the literature examining the associa­ Classes attended included core courses in each year of study within
tion between sleep quality and anxiety among the postsecondary pop­ each faculty. These classes were chosen to ensure that all students in
ulation found that poor sleep quality is associated with increased odds of their respective faculties were informed about the study. In the second
experiencing anxiety globally [33]. However, to our knowledge no wave of recruitment, class instructors sent standard messages to students
research has been done examining the association of poor sleep quality via OTU’s online educational platform. The message contained details
and anxiety within a Canadian context [34–37]. Overall, the studies about the study and provided a link where students could access the
found in the systematic review suggest that poor sleep quality is asso­ survey. This was done to ensure that students who were not in class at
ciated with anxiety in their respective student population, however the time of the in-class presentation (wave 1) were provided the op­
methodological issues may limit the internal validity of these studies portunity to participate. In the third wave of recruitment the Dean of
[34–38]. First, several studies used recruitment strategies (e.g., snow­ each faculty (FHS and FEd) sent a standard message to all the students
balling techniques or recruiting only students enrolled in a psychology and invited them to participate by clicking on a link where the students
class) that may have led to selection bias [39–43]. Second, some studies could access the survey.
did not consider the impact of covariates on the association between Participants completed the survey package online through Google
poor sleep quality and anxiety such as age, academic program, or bio­ forms. Completed surveys were de-identified by OTU’s Information
logical sex [40,41,44-46]. Therefore, the purpose of this study was to Technology (IT) department to maintain the anonymity of each partic­
investigate the association between poor sleep quality and moderate to ipant. All data from completed surveys were securely stored in the
extremely severe anxiety of students within a Canadian context ac­ university’s Google Drive.
counting for covariates.
CMCC
Methods The study was advertised to students at CMCC using social media
(Facebook and Instagram). An email was also sent out to all students
Study design and study population inviting them to take part in the study. Recruitment for first, second, and
third year students occurred during mandatory classes in October and
Prior to commencing this study, research ethics board approval was November. Fourth year interns were recruited during clinical rounds in
obtained from each of the participating institutions (OTU-REB #14,515; September and October. Two separate recruitment periods were selected
CMCC-REB #1709B03). In 2017, a team of investigators designed and because of differences in the academic schedule of the different years of
conducted a cross-sectional study of student enrolled in the Faculty of study. Prior to the in-class administration of the questionnaire, members
Health Sciences (FHS) and Faculty of Education (FEd) at Ontario Tech of the research team attended classes and explained the study to students
University (OTU) (formally known as the University of Ontario Institute using a standardized script. Students who were not in class on the day
of Technology [UOIT]) and the Canadian Memorial Chiropractic College when the survey was administered were given a link through the class
(CMCC) in Ontario, Canada. messaging system and invited to participate online.
OTU is a public postsecondary institution located in Oshawa, Participants provided informed consent and completed the online
Ontario. It offers several undergraduate and post-graduate degrees questionnaire using their personal electronic devices through the stu­
within multiple departments. The FHS offers six programs: Allied Health dent course portal. Completed surveys were de-identified by CMCC’s IT

2
M. Albrecht-Bisset et al. Sleep Epidemiology 3 (2023) 100062

department to maintain the anonymity of each participant. All data from In this survey, food insecurity was measured using the Household Food
completed surveys were securely stored in a Google Drive. Security Survey Model (HFSSM) [54]. The HFSSM is commonly used to
classify participants as either food secure, food insecure, or severely food
Data collection insecure and has been found to have acceptable psychometric properties
[54,55].
An online survey questionnaire was developed through a review of Substance abuse was measured using the Alcohol, Smoking, and
the literature and in consultation with subject matter experts at OTU and Substance Involvement Screening Test (ASSIST). This questionnaire has
CMCC. The survey instrument was designed to measure the presence of been found to have acceptable psychometric properties [56,57]. The
symptoms of depression, anxiety, and stress, as well as the impact on ASSIST measures both the 3-month and lifetime use of tobacco, alcohol,
function and modifiable lifestyle factors. The questionnaire included cannabis, cocaine, amphetamine-type stimulants, inhalants, sedatives,
valid and reliable questions and instruments to measure sleep quality, hallucinogens, opioids, and other drugs [56].
physical activity, substance use, food insecurity, spinal pain and socio- Neck and low back pain intensity was measured with the Musculo­
demographic variables. For the purpose of the present research objec­ skeletal Pain Intensity – Numeric Rating Scale (NRS) [58,59]. The NRS is
tive, two outcome measures were used: sleep quality and anxiety. a horizontal numerical scale that ranges from 0 – 10 with higher scores
indicating more intense pain. The NRS has adequate construct validity
Independent variable: sleep quality and high test-retest reliability [58,59].
Sleep quality was measured using the validated Pittsburgh Sleep Physical activity and sedentary behavior were measured using the
Quality Index (PSQI), which assesses patterns of sleep dysfunction over Physical Activity and Sedentary Behaviour Questionnaire (PASB-Q)
the previous one-month period [47]. The PSQI includes 19 questions [60]. The PASB-Q distinguishes between individuals with either poor,
classified into seven subscales: sleep latency (i.e., how long it usually fair, good, very good, or excellent physical activity. It also has accept­
takes to fall asleep), sleep duration (i.e., total amount of sleep), use of able psychometric properties [60].
sleep medication (i.e., prescribed or over the counter), perceived sleep Finally, sociodemographic information and medical comorbidities
quality (i.e., an individual’s personal rating of their sleep quality), were measured using questions from the Canadian Community Health
habitual sleep efficiency (i.e., percentage of time spent sleeping while in Survey and Statistics Canada. In addition, information pertaining to
bed), sleep disturbances (i.e., factors that cause wakening during the faculty, program, age, annual income, household income, marital status,
night or early morning), and daytime dysfunction (i.e., inability to living situations, ethnicity, dependents, academic average, year of
perform daytime activities such as driving, eating meals or engaging in study, domestic status, social support, employment of primary and
social activities due to sleepiness) [47]. Each item is scored on a 0–3 secondary guardians, and other demographic variables were also
Likert scale, where 3 reflects the negative extreme [47]. The scores of collected.
the seven subscales are summed to generate a global index score ranging
from 0 to 21. A global score >5 is indicative of a poor-quality sleeper, Statistical analysis
while a score of ≤5 is indicative of a good-quality sleeper. The PSQI is a
reliable and valid tool with a high internal consistency (Cronbach’s All analyses were conducted using Statistical Program for the Social
alpha 0.83) [47]. The PSQI has been found to have valid and acceptable Sciences (SPSS) version 27.0 [61]. Descriptive statistics were presented
psychometric properties across a range of different samples including as means (standard deviation) for continuous variables or frequencies
postsecondary students, as well as adequate validity and reliability [13, (percentage) for categorical variables. Participation bias was assessed by
36,38,47-49]. Originally the scoring algorithm recommended a cut-off comparing characteristics of our sample to the characteristics of stu­
score of 5 for indicating good/poor quality sleep (0–5/21 indicates dents enrolled in the FHS, FEd or CMCC respectively. The prevalence of
good sleep quality and 6–21/21indicates poor sleep quality) [47]. moderate to extreme anxiety and poor sleep quality within these sam­
However, increasing the cut-off point to 6/21 increased the sensitivity of ples was measured by dividing the number of participants with the
the PSQI to 100% to distinguish between individuals with insomnia and outcome (numerator) by the total sample size (denominator).
individuals without insomnia [50]. Therefore, in this study the scores Logistic regression models were used to describe the association
will be dichotomized with 0–6/21 indicating good sleep quality and between sleep quality and moderate to extremely severe anxiety. Two
7–21/21 indicating poor sleep quality. separate sets of regression models were built. In the first set of models,
the PSQI scores were dichotomized as good quality and poor sleep
Dependent variable: anxiety quality categories. In the second set, the PSQI scores were analyzed as a
The Depression Anxiety Stress Scale – 21 (DASS-21) is the short form continuous variable.
of the DASS-42, which is used to measure symptoms of depression, The regression models were each built using a three-step process.
anxiety and stress in the past week [51]. It includes 21 questions and First, a bivariate model was built to determine the crude association
produces scores for three different subscales: depression, anxiety and between sleep quality and anxiety. Second, separate multivariate
stress [51]. The DASS-21 has been found to be valid and reliable in many models were built with potential covariates to assess their impact on the
populations including postsecondary students [6,52]. The anxiety sub­ association between poor sleep quality and moderate to extremely se­
scale is scored from 0 to 42 and further categorized into five levels; vere anxiety. The following covariates were considered in the models
normal, mild, moderate, severe, and extremely severe [51]. Anxiety that based on the literature and in consultation with subject experts: food
is categorized as moderate to extremely severe on the DASS-21 is also insecurity, physical activity, comorbidities, prevalence of substance
classified as clinically significant anxiety, which is being investigated in abuse, neck and low back pain, trouble coping, depression, stress, age,
this study [51,52]. The anxiety subscale from the DASS-21 has shown to gender, biological sex, program, ethnicity, residency status, marital
have high internal consistency with an overall reliability (Cronbach’s status, personal annual income, household annual income, living ar­
alpha) of 0.851. rangements, and number of hours worked per week. If a particular
variable changed the association (beta) between sleep quality and anx­
Covariates iety by 10% or more, it was considered an important covariate. Finally, a
The survey instrument included other questionnaires to measure multivariate model was built that included all the covariates that met
food insecurity, substance abuse, neck and low back pain, physical ac­ the criteria described above. The association between poor sleep quality
tivity, and sociodemographic information. and moderate to extremely severe anxiety is reported as odds ratios (OR)
Food insecurity is often defined as a disruption in eating patterns or a and 95% confidence intervals (CI).
lack of food intake due to insufficient money and other resources [53]. Data from both faculties at OTU and CMCC were explored to

3
M. Albrecht-Bisset et al. Sleep Epidemiology 3 (2023) 100062

determine if the data could be pooled by comparing the descriptive Table 1

statistics and crude measures of associations between poor sleep quality Baseline characteristics of study sample.
and moderate to extremely severe anxiety within the two faculties. Characteristics FHS FEd OTU CMCC
Additionally, the Z scores and corresponding p-values were computed to (n = (n = combined (n =
determine if logistic regression coefficients differed significantly 675) 207) (n = 882) 510)
(p<0.05) across faculties and institutions. It was found that the two data Age, Mean (SD) 22.1 25.6 22.9 (5.5) 24.6
sets from OTU could be pooled together because the descriptive statistics (5.5) (4.8) (2.8)
and crude measures of associations between poor sleep quality and Gender
Female 539 141 680 (77.0) 307
moderate to extremely severe anxiety within the two faculties were Male (80) (69.1) 193 (21.8) (60.2)
similar. The data from CMCC could not be pooled with the OTU sample 128 65 203
because of slight differences in population and data collection methods. (19.0) (31.4) (39.8)
Year of Study* n (%)
1 171 69 240 (27.2) 131
Results
2 (25.3) (33.3) 245 (27.7) (25.7)
3 173 72 142 (16.0) 137
Participation rate and sample characteristics 4 (25.6) (34.8) 179 (20.2) (26.9)
5+ 142 0 76 (8.6) 149
The participation rate for students at OTU was 40.1% (882/2199); (21.0) 0 (29.2)
179 66 93
675 participants were from the FHS (35% participation rate) and 207 (25.6) (31.9) (18.2)
participants from the FEd (77% participation rate). Of the total sample 10 (1.4) 0
(FHS and FEd combined), the mean age was 22.9 (SD 5.5) and 77% of Cultural/Ancestral
the participants were female (Table 1). For participants at CMCC the Ethnicity* 35 (5.2) 5 (2.4) 40 (4.5) 15 (2.9)
Aboriginal/First Nations/ 58 (8.5) 8 (3.8) 66 (7.4) 10 (2.0)
participation rate was 67% (510/766). The sample mean age was 25
Metis 349 126 415 (53.8) 381
years (SD: 2.8) and 60% of the participants were female (Table 1). Black (51.7) (60.9) 130 (14.7) (74.7)
Caucasian 105 25 123 (12.9) 47 (9.2)
Prevalence of poor sleep quality and anxiety East Asian (15.5) (12.1) 67 (7.5) 23 (4.5)
South Asian 105 18 (8.6) 64 (7.2) 16 (3.1)
South East Asian (15.5) 14 (6.7) 37
The one-month prevalence of poor sleep quality (scores ≥7/21 on Latin American/Middle 53 (7.8) 12 (5.8) (7.25)
the global PSQI) among Ontario students was 61.8% (95% CI: 9.5 – 9.9). Eastern 52 (7.7)
The mean PSQI score was 7.7/21 (95% CI: 7.5 – 8.0), suggesting that the Annual Personal Income*
majority of participants in this sample were poor-quality sleepers. $0 - $4999 351 85 406 (46.0) 311
$5000 - $9999 (52.5) (41.1) 248 (28.1) (61.0)
Additionally, 73.7% of students were obtaining less than 7 h of sleep per
$10,000 - $19,999 184 64 130 (14.7) 129
night and 41.7% reported experiencing daytime dysfunction (including Above $20,000 (27.5) (30.9) 89 (10.0) (25.3)
excessive daytime sleepiness and trouble staying awake during tasks 94 36 51
such as eating, driving and socializing with friends). In the sample of (14.1) (17.4) (10.0)
CMCC students, the one-month prevalence of poor sleep quality was 69 20 (9.7) 16 (3.1)
(10.2)
59.8% (95% CI 9.2 – 9.8). The mean PSQI score was 7.4/21 (95% CI 7.1 Annual Household Income*
– 7.7), with just over half of the students classified as poor-quality $0 - $49,000 159 42 201 (22.8) 244
sleepers. Additionally, 75.5% of students report obtaining less than 7 $50,000 - $59,000 (23.5) (20.3) 136 (15.4) (47.8)
h of sleep per night and 32.4% report daytime dysfunction. The one- $60,000 - $79,000 103 33 143 (16.2) 40 (7.8)
Above $80,000 (15.2) (15.9) 385 (43.6) 52
week prevalence of reporting moderate to extremely severe anxiety in
104 39 (11.2)
OTU students was 58.3% (95% CI 14.9 – 16.1), with the mean DASS (15.4) (18.8) 164
score of 10.4/21 (95% CI 9.8–10.9). Furthermore, a statistically signif­ 300 85 (32.2)
icant correlation between PSQI scores and DASS-21 anxiety scores (r = (44.4) (41.1)
0.30, p<0.001) was found. For CMCC students, the one-week prevalence Marital Status*
Single, never married 601 172 773 (87.6) 451
of reporting moderate to extremely severe anxiety was 41.0% (95% CI Married/Common Law (89.0) (83.1) 91 (10.3) (88.4)
14.6 – 16.5 and the mean DASS score was 7.4/21 (95% CI 6.8 – 8.0). A 59 (8.7) 32 59
statistically significant correlation between PSQI scores and DASS-21 (15.4) (11.6)
anxiety scores (r = 0.32, p<0.001) was also found in this sample. Parents Marital Status
Single 34 (5.0) 9 (4.3) 43 (4.9) 33 (6.5)
Married/Common Law 125 38 163 (18.5) 378
Association between sleep quality and anxiety Separated/Divorced (18.5) (18.4) 644 (73.0) (74.1)
495 149 86
The separate regression models that were built for the FHS and FED (73.3) (72.0) (16.9)
and revealed that the point estimates (betas) were similar for both fac­ Employment Status of
Primary* Guardian 525 129 654 (74.1) 372
ulties (Table 2 and Table 3), therefore the data from both faculties was Full-time (77.8) (62.3) 50 (5.6) (72.9)
pooled. Data from CMCC could not be pooled given the differences in Part-time 42 (6.2) 8 (3.9) 38 (4.1) 36 (7.1)
data collection and population characteristics Therefore, the data for Retired 21 (3.1) 17 (8.2) 56
CMCC and OTU will be reported separately. (11.0)
Hours of Paid Work Per
In the first model where the PSQI was treated as a dichotomous
Week* 241 77 318 (36.0) 263
variable, the crude analysis suggested that OTU and CMCC students who 0 (35.7) (37.2) 393 (44.5) (51.6)
were classified as poor-quality sleepers (PSQI score ≥7/21) were more 1–19 h 294 99 171(19.3) 229
likely to report moderate to extreme anxiety (OR = 3.2 [95% CI 2.4 – 4.3 20+ hours (43.5) (47.8) (44.9)
and OR = 3.8 [95% CI 2.5 – 5.9] respectively). After controlling for 140 31 18 (3.5)
(20.6) (13.0)
relevant covariates (biological sex, depression and problems handling Commute Time to
stress) the odds ratios decreased to 1.5 (95% CI 1.1 – 2.2) for OTU and University* 244 36 280 (31.7) 258
2.6 (95% CI 1.6 – 4.2) for CMCC (Table 2 and Table 3). (continued on next page)
In the second model, where the PSQI was treated as a continuous

4
M. Albrecht-Bisset et al. Sleep Epidemiology 3 (2023) 100062

Table 1 (continued ) Table 2

Characteristics FHS FEd OTU CMCC
Association between poor sleep quality (PSQI score*) and clinically significant
(n = (n = combined (n = anxiety (DASS-21 Anxiety score**) – OTU.
675) 207) (n = 882) 510) Odds Ratio (95% CI) of reporting
Less than 15 Min (36.1) (18.3) 196 (22.2) (50.6) moderate-extreme anxiety
15 – 19 Min 158 38 123 (13.9) 113 Crude Adjusted
30 – 44 Min (23.4) (18.8) 283 (32.0) (22.2) Analysis Analysis†
45+ Minutes 93 30 56
(13.8) (14.4) (11.0) PSQI Dichotomous (poor sleep quality vs. good 3.3 (2.4 – 1.6 (1.1 – 2.2)
180 103 83 sleep quality) 4.3)
(26.7) (49.7) (16.3) PSQI Continuous 1.3 (1.2 – 1.1 (1.0 – 1.2)
Living Situation* 1.3)
Relatives 435 150 585 (66.3) 152 *
PSQI (Pittsburgh Sleep Quality Index) = Good sleep quality ≤ 6, Poor sleep
Roommate/housemate (66.4) (72.5) 118 (13.4) (29.8)
quality ≥ 7;.
Student Residence 106 12 (5.8) 84 (9.5) 188 **
Living with a Partner (15.7) 4 (1.9) 99 (11.2) (36.9) DASS-21 (Depression Anxiety Stress Scale – 21 Anxiety subscale) = Not
80 33 45 (8.8) clinically significant anxiety (no or mild anxiety) ≤ 9 Clinically significant
(11.9) (15.9) 95 anxiety (moderate to extremely severe anxiety) ≥ 10.
66 (9.8) (18.6) †
covariates adjusted for were biological sex, depression and problems
Academic Average in handling stress.
Previous Year* 80 3 (1.45) 83 (9.4) 8 (1.6)
Between 60 – 69% (11.8) 34 265 (30.0) 114
Between 70 – 79% 231 (16.4) 390 (44.2) (22.4)
Table 3
Between 80 – 89% (34.0) 126 124 (14.0) 312
Between 90 – 100% 264 (60.9) (61.2) Association between poor sleep quality (PSQI score*) and clinically significant
(39.1) 44 76 anxiety (DASS-21 Anxiety score**) – CMCC.
80 (21.2) (14.9) Odds Ratio (95% CI) of reporting
(11.8) moderate-extreme anxiety
Number of Dependents*
None 490 162 652 (73.9) 457 Crude Adjusted
1–6+ (72.6) (78.2) 230 (26.1) (89.6) Analysis Analysis†
185 45 53 PSQI Dichotomous (poor sleep quality vs. good 3.8 (2.5 – 2.6 (1.6 – 4.2)
(27.4) (21.7) (10.4) sleep quality) 5.9)
Diagnosed Medical PSQI Continuous 1.3 (1.2 – 1.2 (1.1 – 1.3)
Conditions* 196 60 256 (29.0) 128
1.4)
Allergies (29.0) (29.0) 120 (13.6) (25.1)
*
Asthma 98 22 73 (8.3) 62 PSQI (Pittsburgh Sleep Quality Index) = Good sleep quality ≤ 6, Poor sleep
Migraines (14.5) (10.6) (12.2) quality ≥ 7;.
61 (9.0) 12 (5.8) 38 (7.5) **
DASS-21 (Depression Anxiety Stress Scale – 21 Anxiety subscale) = Not
Neck Pain* clinically significant anxiety (no or mild anxiety) ≤ 9 Clinically significant
No 245 94 339 (38.4) 118
anxiety (moderate to extremely severe anxiety) ≥ 10.
Yes (36.3) (45.4) 543 (61.6) (23.1)
430 113 392
†
covariates adjusted for were biological sex, depression and problems
(63.7) (54.6) (76.9) handling stress.
Low Back Pain*
No 220 81 301 (34.1) 158 variable, the crude analysis showed that poor quality sleepers were also
Yes (32.6) (39.1) 581 (65.9) (31.0)
455 126 352
more likely to report moderate to extreme anxiety (OTU: OR = 1.2 [95%
(67.4) (60.9) (69.0) CI 1.2 – 1.3]; CMCC: (OR = 1.3 [95% CI 1.2 – 1.4]). After controlling for
Lifetime Substance Use* important covariates (biological sex, depression and problems handling
Tobacco 226 87 313 (35.5) 185 stress) the association decreased (OTU: OR = 1.1 [95% CI 1.0 – 1.2] and
Alcohol (33.5) (42.0) 706 (80.0) (36.3)
CMCC: (OR = 1.2 [95% CI 1.1 – 1.3]) (Table 2 and Table 3).
Marijuana 528 178 404 (45.8) 481
Cocaine (78.2) (86.0) 51 (5.7) (94.3)
Stimulants 298 106 82 (9.3) 277 Discussion
Sedatives (44.1) (51.2) 69 (7.8) (54.3)
Hallucinogens 35 (5.2) 16 (7.7) 58 (6.6) 41 (8.0) The purpose of this study was to assess the association between poor
62 (9.2) 20 (9.7) 79
50 (7.4) 19 (9.2) (15.5)
sleep quality and moderate to severe anxiety in postsecondary students
40 (5.9) 18 (8.7) 32 (6.3) in Canada. Students from two faculties at OTU and students at CMCC
54 reported a high prevalence of poor sleep quality.
(10.6) The one-month prevalence of poor sleep quality was high in both the
Food Insecurity*
OTU (61.8%) and CMCC (59.8%) sample population despite using
Yes 189 47 236 (27.5) 70
No (29.1) (22.7) 621 (72.5) (13.7) conservative cut-off scores on the PSQI. Since we used a higher and more
461 160 440 conservative cut-off score to differentiate between good and poor sleep
(70.9) (77.3) (86.3) quality this may be an underestimation of poor sleep quality within this
Sleep Quality* population. Poor sleep quality is not unique to the students in our
Good 240 79 314 (35.6) 225
Poor (35.6) (38.2) 554 (62.8) (44.1)
sample. These results are consistent with findings in samples of students
435 128 285 from Saudi Arabia, Iran, Egypt, Nigeria, Brazil, Australia, Egypt and the
(64.4) (61.8) (55.9) United States where similar proportions of students reported experi­
Anxiety* encing poor sleep quality (one-month prevalence ranged from 48.5% –
Normal/Mild 347 118 368 (41.7) 301
63.2%) [36-38,40,43,48,62,63].
Moderate/Severe/Extreme (51.4) (57.0) 514 (58.3) (59.0)
328 89 209 In addition to the high prevalence of poor sleep quality, it was also
(48.6) (43.0) (41.0) found that the majority of students at both OTU (73.7%; n = 667) and
* CMCC (75.5%; n = 385) were obtaining insufficient sleep duration (<7 h
n (%).
of sleep per night). The National Sleep Foundation recommends that

5
M. Albrecht-Bisset et al. Sleep Epidemiology 3 (2023) 100062

emerging adults (between 19 and 25 years of age) obtain 7–9 h of sleep such as faculty, counselling centres, campus health centres and policy
per night [3,18,64]. This is problematic since students who obtain makers. Sleep difficulties are highly prevalent among postsecondary
insufficient sleep are more likely to experience cognitive (such as student populations globally. Evidence of the association between poor
memory issues), emotional (such as increased impulsivity and poor sleep and anxiety may help the wider postsecondary community to
mood regulation) and physical (impaired immune function, excessive develop policies to help minimize anxiety in students by targeting sleep.
daytime sleepiness) impairments [20,39,46,65-69]. However, more research is needed to understand if poor sleep quality is
Additionally, this study found that 58.3% of OTU and 41.0% of a risk factor for anxiety within this population. Many small changes can
CMCC students reported moderate to extremely severe anxiety in the be implemented within the postsecondary community. For example,
past week. This is similar to the high proportions of postsecondary faculty can make changes to their evaluative methods (such as changing
students in Saudi Arabia, Brazil, Australia, Ethiopia, Thailand and Iran assignment due dates to midnight instead of early morning to discourage
who also reported significant anxiety symptoms [35-38,40,70]. The students from staying up all night to complete assignments), institutions
number of students reporting clinically significant anxiety globally is can change course timetables to minimize early morning or late-night
increasing [39,44,52,71]. Therefore, more studies are needed to inves­ classes, and campus health and counselling centres can run informa­
tigate potential risk factors associated with anxiety and to identify cor­ tion sessions for students on sleep hygiene and the importance of sleep.
relates, which is an important early step towards the goal of reducing the
prevalence of anxiety [4]. Strengths and limitations
The present study found that students who were classified as poor-
quality sleepers were more likely to report anxiety when compared to This study has many strengths. First, sleep quality and anxiety were
those who were good quality sleepers even after controlling for cova­ measured using valid and reliable instruments that have been used
riates (depression, biological sex and problems handling stress) sug­ among postsecondary students which reduces the potential for mea­
gesting sleep may have important protective factor in the development surement bias. Second, the participation rate was high in both FEd
of anxiety. These findings are consistent with other studies of under­ students (77%) and residents at CMCC (67%). Third, a wide range of
graduate students conducted in a range of countries such as Australia, variables were assessed as potential covariates, and this allowed the
Thailand, the UK, Iran, the US and China [35,37,39,41,49,62,70,72-74]. final model to be adjusted for the variables that were determined to be
Significant associations were also found in studies sampling graduate important. Potential covariates that were considered in this model were
and professional students [34,38,40,42,44,75]. identified through the literature and consultations from subject matter
The results of this study are consistent with the results found in other experts. Finally, to our knowledge this is one of the first studies inves­
studies conducted in a variety of different countries [13,37,41,76,77, tigating the relationship between sleep quality and anxiety in a Cana­
78]. While previous research has found sleep quality and anxiety to be dian sample of postsecondary students.
associated, the current study examined this association by controlling This study also has a number of limitations. Selection bias may have
for a variety of covariates in the regression analysis which previous been present in this study as the participation rate in the FHS was 35%.
study have not done [35,37,39-42,45,48,49,62,68,70,73,74,76-86]. It However, the characteristics of our sample in the FHS closely represent
was found that depression, stress and biological sex were important the characteristics of the general FHS population, which reduces the
covariates in the association between sleep and anxiety in postsecondary likelihood of participation bias. Second, although online recruitment
students. While these covariates have been explored in other studies strategies were incorporated into the study, the majority of students who
involving the general population there have been no studies to our participated were recruited through in-class recruitment (first wave of
knowledge examining these covariates within the postsecondary student recruitment). Therefore, the data from students who did not attend class
population, thus making this study novel [6,7,33,79,80]. There is during this stage of recruitment were not included. Since we have no
growing recognition and concern that postsecondary students are a way of knowing if there were any differences between the students who
separate group at risk for both significant sleep problems and mental were in class during recruitment and those who were not, this may have
health issues (including depression, anxiety and suicidal ideation) [3,4, potentially biased our results.
39,44,78]. As postsecondary students are reporting a higher prevalence
of mental health issues and sleep disturbances compared to the general Conclusion
population it is important that future research continue to recognize this
group as a vulnerable population and separate them out as opposed to Poor sleep quality and anxiety are both prevalent among post­
clustering them with the general population or emerging adults [3,4, secondary students. A positive association between poor sleep quality
14]. and moderate to extremely severe anxiety was found among a sample of
Although this study provides further evidence of an association be­ students enrolled in the Faculty of Health Sciences and the Faculty of
tween poor sleep quality and anxiety, no causal links can be made. Education at Ontario Tech University as well as residents at the Cana­
Future cohort studies are needed to investigate if poor sleep quality is a dian Memorial Chiropractic College. These findings suggest the impor­
risk factor for the development of anxiety in postsecondary students. tance of promoting healthier sleep habits which could impact mental
Three previous studies (two in the United States and one in China) have health outcomes including anxiety. However, further studies are needed
found that students who report poor sleep quality at baseline are more to investigate sleep quality as a risk factor for the development of clin­
likely to report anxiety at follow-up (at 3 months, 6 months and 1-year ically significant anxiety. The results of this study provide information
follow-up) [34,49,63]. Two of these studies reported on very specific that is useful both to postsecondary students as well as the broader
populations (one on only male students and one on medical students) postsecondary community.
and therefore more research needs to be done to see if this finding can be
replicated in different student populations [34,49]. Understanding the Funding
directionality of the association has the potential to help create impor­
tant options to prevent (or reduce) anxiety among students. If poor sleep This research was supported by Ontario Trillium Foundation
quality is found to be a risk factor in the development of anxiety, (SD97818).
treatments and interventions aimed at improving sleep quality could
possibly reduce the likelihood of developing clinically significant Declaration of Competing Interest
anxiety.
The findings from this study have important implications not only for The authors declare that they have no known competing financial
postsecondary students but also for the wider postsecondary community interests or personal relationships that could have appeared to influence

6
M. Albrecht-Bisset et al. Sleep Epidemiology 3 (2023) 100062

the work reported in this paper. [32] Ohayon M, M, Roth T. What are the contributing factors for insomnia in the general
population? J Psychosom Res 2001;51:745–55.
[33] Albrecht-Bisset M, et al. The association between sleep quality and anxiety in
Acknowledgments postsecondary students: a systematic review of the literature. Sleep Med 2022. in
publication - March 2022.
We would like to acknowledge our partners at the Canadian Mental [34] Kalmbach DA, et al. Insomnia symptoms and short sleep predict anxiety and worry
in response to stress exposure: a prospective cohort study of medical interns. Sleep
Health Association (CMHA) Durham Region as well as at Ontario Shores Med 2019;55:40–7.
centre for Mental Health Sciences. We also acknowledge Ontario Tech [35] Pensuksan WC, et al. Relationship between Poor Sleep Quality and Psychological
University (formerly known as the University of Ontario Institute of Problems among Undergraduate Students in the Southern Thailand. Walailak
Journal of Science and Technology 2016;13(4):11.
Technology); Institute for Disability and Rehabilitation Research [36] Rezaei M, et al. Sleep quality and its association with psychological distress and
(IDRR); and the Canadian Memorial Chiropractic College (CMCC). sleep hygiene: a cross-sectional study among pre-clinical medical students. Sleep
Sci 2018;11(4):274–80.
[37] Di Benedetto M, Towt CJ, Jackson ML. A Cluster Analysis of Sleep Quality, Self-
References Care Behaviors, and Mental Health Risk in Australian University Students. Behav
Sleep Med 2020;18(3):309–20.
[1] Glauser W. Postsecondary campuses responding to record anxiety and depression [38] Carvalho Aguiar Melo M, et al. Sleep quality among psychiatry residents. Can J
levels. CMAJ 2017;189(48):E1501–2. Psychiatry 2016;61(1):44–9.
[2] Smetanin, P., Stiff, D., Briante, C., Adair, C.E., Ahmad, S., & Khan, M., The life and [39] Adams SK, Kisler TS. Sleep quality as a mediator between technology-related sleep
economic impact of major mental illnesses in Canada 2011-2041. 2011. quality, depression, and anxiety. Cyberpsychol Behav Soc Netw 2013;16(1):25–30.
[3] Council Of Ontario Universities. It together: taking action on student mental [40] Al-Khani AM, et al. A cross-sectional survey on sleep quality, mental health, and
health. 2017. p. 12. academic performance among medical students in Saudi Arabia. BMC Res Notes
[4] American College Health Association, American college health association-national 2019;12(1):665.
college health assessment II: canadian consortium executive summary spring 2019. [41] Carciofo R. Morning affect, eveningness, and amplitude distinctness: associations
2019. with negative emotionality, including the mediating roles of sleep quality,
[5] Children’s Mental Health Ontario, Kids can’t wait: 2020 report on wait lists and personality, and metacognitive beliefs. Chronobiol Int 2020;37(11):1565–79.
wait times for child and youth mental health care in Ontario. 2020. p. 12. [42] Rahimian Boogar I, Mirkouhi MGhodrati. An investigation of factors influencing
[6] Beiter R, et al. The prevalence and correlates of depression, anxiety, and stress in a disrupted sleep in university students. Pars of Jahrom University of Medical
sample of college students. J Affect Disord 2015;173:90–6. Sciences 2013;11(4):25–33.
[7] Choi HY, et al. A Study on Correlation between Anxiety Symptoms and Suicidal [43] Ferraro FR, et al. Texting/iPod dependence, executive function and sleep quality in
Ideation. Psychiatry Investig 2011;8(4):320–6. college students. Comput Hum Behav 2015;49:44–9.
[8] Eller T, et al. Symptoms of anxiety and depression in Estonian medical students [44] Choueiry N, et al. Insomnia and relationship with anxiety in university students: a
with sleep problems. Depress Anxiety 2006;23(4):7. cross-sectional designed study. PLoS ONE 2016;11(2):e0149643.
[9] Lovell GP, et al. A cross-sectional investigation of depressive, anxiety, and stress [45] Wong ML, et al. The interplay between sleep and mood in predicting academic
symptoms and health-behavior participation in Australian university students. functioning, physical health and psychological health: a longitudinal study.
Nurs Health Sci 2015;17(1):134–42. J Psychosom Res 2013;74(4):271–7.
[10] Schlarb AA, Kulessa D, Gulewitsch MD. Sleep characteristics, sleep problems, and [46] Ye YL, et al. Associations between multiple health risk behaviors and mental health
associations of self-efficacy among German university students. Nat Sci Sleep 2012; among Chinese college students. Psychol Health Med 2016;21(3):377–85.
4. [47] Buysse DJ, et al. The Pittsburgh sleep quality index: a new instrument for
[11] Storrie K, Ahern K, Tuckett A. A systematic review: students with mental health psychiatric practice and research. Psychiatry Res 1989;2(28):193–213.
problems–a growing problem. Int J Nurs Pract 2010;16(1):1–6. [48] Fawzy M, Hamed SA. Prevalence of psychological stress, depression and anxiety
[12] Gaultney JF. The prevalence of sleep disorders in college students: impact on among medical students in Egypt. Psychiatry Res 2017;255:186–94.
academic performance. J Am College Health 2010;59(2):7. [49] Zou P, et al. Poorer sleep quality correlated with mental health problems in college
[13] Lund HG, et al. Sleep patterns and predictors of disturbed sleep in a large students: a longitudinal observational study among 686 males. J Psychosom Res
population of college students. J Adolesc Health 2010;46(2):124–32. 2020;136:110177.
[14] Owens J. Insufficient sleep in adolescents and young adults: an update on causes [50] Backhaus J, et al. Test–retest reliability and validity of the Pittsburgh sleep quality
and consequences. Pediatrics 2014;134(3):e921–32. index in primary insomnia. J Psychosom Res 2002;53(3):4.
[15] Papaconstantinou EA, et al. The Association Between Chronotype and Mental [51] Lovibond SH, Lovibond PF. Manual for the depression anxiety stress scales.
Health Problems in a University Population: a Systematic Review of the Literature. Psychology Foundation of Australia; 1995.
Int J Ment Health Addict 2018;17(3):716–30. [52] Ahmad N, et al. The validity and reliability of psychometric profile for depression,
[16] Eliasson A, et al. Association of seelp and acadmeic performance. Sleep and anxiety and stress scale (dass21) instrument among malaysian undergraduate
Breathing 2002;6:4. students. Int J Acad Res Bus Soc Sci 2018;8(6).
[17] Gomes AA, Tavares J, de Azevedo MH. Sleep and academic performance in [53] Coleman-Jensen A, GC, Singh A. Household food security in the United States in
undergraduates: a multi-measure, multi-predictor approach. Chronobiol Int 2011; 2014. USDA-ERS Econ Res Rep 2015;194.
28(9):786–801. [54] Bickel G, et al. Guide to measuring household food security revised USDA-FNS
[18] Holmes A, Silvestri R. Rates of mental illness and associated academic impacts in office of analysis. Nutr Eval 2000;(6).
ontario’s college students. Canadian J School Psychol 2015;31(1):27–46. [55] Gulliford MD MC, Rocke B. Reliability and validity of a short form household food
[19] Nyer M, et al. Relationship between sleep disturbance and depression, anxiety, and security scale in a Caribbean community. BMC Public Health 2004;4(22).
functioning in college students. Depress Anxiety 2013;30(9):8. [56] Humeniuk R, Ali R. World health organization, validation of the alcohol, smoking
[20] Babson KA, et al. An examination of anxiety sensitivity as a moderator of the and substance involvement screening test (ASSIST) and pilot brief intervention: a
relation between sleep anticipatory anxiety and sleep onset latency. J Cogn technical report of phase ii findings of the who assist project. World Health
Psychother 2008;22(3):258–70. Organization; 2006.
[21] Puzino K, et al. Am I (hyper)aroused or anxious? Clinical significance of pre-sleep [57] Newcombe DA, Humeniuk RE, Ali R. Validation of the world health organization
somatic arousal in young adults. J Sleep Res 2019;28(4):e12829. alcohol, smoking and substance involvement screening test (ASSIST): report of
[22] Bandelow B, Michaelis S. Epidemiology of anxiety disorders in the 21st century. results from the Australian site. Drug Alcohol Rev 2005;3(24):217–26.
Dialogues clin neurosci 2015;3(17):8. [58] Childs JD, Piva SR, Fritz JM. Responsiveness of the numeric pain rating scale in
[23] Martin EI, et al. The neurobiology of anxiety disorders: brain imaging, genetics, patients with low back pain. Spine 2005;11(30):1331–4.
and psychoneuroendocrinology. Psychiatr Clin North Am 2009;32(3):549–75. [59] Gallasch CH, Alexandre N. The measurement of musculoskeletal pain intensity: a
[24] Prabha L, et al. A comparative study of health anxiety in neurology and psychiatry comparison of four methods. Revista gaúcha de enfermagem/EENFUFRGS 2007.
settings. J Neurosci Rural Pract 2020;11(1):125–9. [60] Fowles JR, et al. A pilot study: validity and reliability of the CSEP− PATH PASB-Q
[25] American Psychiatric Association, Anxiety disorders, in diagnostic and statistical and a new leisure time physical activity questionnaire to assess physical activity
manual of mental disorders. 2013. and sedentary behaviours. Appl Physiol, Nutr Metabolism 2017;42(6):677–80.
[26] Anwer S, et al. Psychometric analysis of the sleep hygiene index and correlation [61] IBM Corp. IBM spss statistics for windows version 27. Armonk NY: IBM Corp; 2020.
with stress and anxiety among saudi university students. Nat Sci Sleep 2019;11:8. [62] Seun-Fadipe CT, Mosaku KS. Sleep quality and psychological distress among
[27] Deboer T. Sleep homeostasis and the circadian clock: do the circadian pacemaker undergraduate students of a Nigerian university. Sleep Health 2017;3(3):190–4.
and the sleep homeostat influence each other’s functioning? Neurobiol Sleep [63] Doane LD, Gress-Smith JL, Breitenstein RS. Multi-method assessments of sleep over
Circadian Rhythms 2018;5:68–77. the transition to college and the associations with depression and anxiety
[28] Krystal AD, Edinger JD. Measuring sleep quality. Sleep Med 2008;9:S10–7. symptoms. J Youth Adolesc 2015;44(2):389–404.
[29] Zielinski MR, McKenna JT, McCarley RW. Functions and mechanisms of sleep. [64] Prevatt F, et al. The academic success inventory for college students. J College
AIMS Neurosci 2016;3(1):67–104. Admission 2011;(211):5.
[30] Alvaro PK, Roberts RM, Harris JK. A systematic review assessing bidirectionality [65] Silva A, et al. Sleep quality evaluation, chronotype, sleepiness and anxiety of
between sleep disturbances, anxiety, and depression. Sleep 2013;36(7):1059–68. Paralympic Brazilian athletes: beijing 2008 Paralympic Games. Br J Sports Med
[31] Mitchel HA, Weinshenker D. Good night and good luck: norepinephrine in sleep 2012;46(2):150–4.
pharmacology. Biochem Pharmacol 2010;79:9. [66] Sivertsen B, et al. Delayed sleep phase syndrome in adolescents: prevalence and
correlates in a large population based study. BMC Public Health 2013;1163(13):10.

7
M. Albrecht-Bisset et al. Sleep Epidemiology 3 (2023) 100062

[67] Staner L. Sleep and anxiety disorders. Dialogues Clin Neurosci 2003;5(3):10. [78] Dickinson DL, et al. Personal sleep debt and daytime sleepiness mediate the
[68] Zhang Y, Peters A, Chen G. Perceived stress mediates the associations between relationship between sleep and mental health outcomes in young adults. Depress
sleep quality and symptoms of anxiety and depression among college nursing Anxiety 2018;35(8):775–83.
students. Int J Nurs Educ Scholarsh 2018;15(1). [79] Andrews B, Wilding JM. The relation of depression and anxeity to life-stress and
[69] Zochil ML, Thorsteinsson EB. Exploring poor sleep, mental health, and help- achievement in students. British Journal of Psychology 2004;95:12.
seeking intention in university students. Aust J Psychol 2018;70(1):7. [80] Becker SP, et al. Sleep in a large, multi-university sample of college students: sleep
[70] Lemma S, et al. Sleep quality and its psychological correlates among university problem prevalence, sex differences, and mental health correlates. Sleep Health
students in Ethiopia: a cross-sectional study. BMC Psychiatry 2012;12(237):7. 2018;4(2):8.
[71] Javadi A, Haji S, Shafikhani AA. Evaluation of depression and anxiety, and their [81] Doos Ali Vand H, et al. Prediction of insomnia severity based on cognitive,
relationships with insomnia, nightmare and demographic variables in medical metacognitive, and emotional variables in college students. Explore (NY) 2014;10
students. Sleep and Hypnosis 2019;21(1):7. (4):233–40.
[72] Kirwan M, Pickett SM, Jarrett NL. Emotion regulation as a moderator between [82] Hsu CY, et al. Associations between chronotypes, psychopathology, and personality
anxiety symptoms and insomnia symptom severity. Psychiatry Res 2017;254:40–7. among incoming college students. Chronobiol Int 2012;29(4):491–501.
[73] Norbury R, Evans S. Time to think: subjective sleep quality, trait anxiety and [83] Kalyani MN, et al. Investigation of the relationship between psychological
university start time. Psychiatry Res 2019;271:214–9. variables and sleep quality in students of medical sciences. Depress Res Treat 2017.
[74] Ramon-Arbues E, et al. The Prevalence of Depression, Anxiety and Stress and Their [84] Shadzi MR, Salehi A, Vardanjani HM. Problematic Internet Use, Mental Health, and
Associated Factors in College Students. Int J Environ Res Public Health 2020;17 Sleep Quality among Medical Students: a Path-Analytic Model. Indian J Psychol
(19). Med 2020;42(2):8.
[75] Shen Y, et al. Excessive daytime sleepiness in medical students of Hunan province: [85] Taylor DJ, et al. Epidemiology of insomnia in college students: relationship with
prevalence, correlates, and its relationship with suicidal behaviors. J Affect Disord mental health, quality of life, and substance use difficulties. Behav Ther 2013;44
2019;255:90–5. (3):339–48.
[76] Cellini N, et al. Sleep quality and quantity in Italian University students: an [86] Taylor DJ, et al. Insomnia and mental health in college students. Behav Sleep Med
actigraphic study. Chronobiol. Int. 2020:14. 2011;9(2):10.
[77] Tao S, et al. Effects of Sleep Quality on the Association between Problematic
Mobile Phone Use and Mental Health Symptoms in Chinese College Students. Int J
Environ Res Public Health 2017;14(2).