Insomnia and Relationship with Anxiety in University Students: A Cross-Sectional Designed Study

Purpose Sleep disorders (SDs) are now recognized as a public health concern with considerable psychiatric and societal consequences specifically on the academic life of students. The aims of this study were to assess SDs in a group of university students in Lebanon and to examine the relationship between SDs and anxiety. Methods An observational cross-sectional study was conducted at Saint-Joseph University, Lebanon, during the academic year 2013–2014. Four questionnaires were face-to-face administered to 462 students after obtaining their written consent: Insomnia Severity Index (ISI), Pittsburgh Sleep Quality Index (PSQI), Epworth Sleepiness Scale (ESS), and Generalized Anxiety Disorder 7-item scale (GAD-7). Results The prevalence of clinically significant insomnia was 10.6% (95% CI: 7.8–13.4%), more frequent in first year students. ISI mean score was 10.06 (SD = 3.76). 37.1% of the participants were poor sleepers. Excessive daytime sleepiness (EDS) and poor sleep were significantly more frequent among participants with clinical insomnia (p = 0.031 and 0.001 respectively). Clinically significant anxiety was more frequent in students suffering from clinical insomnia (p = 0.006) and in poor sleepers (p = 0.003). 50.8% of the participants with clinically significant anxiety presented EDS versus 30.9% of those with no clinically significant anxiety (p<0.0001). Conclusions The magnitude of SDs in this sample of Lebanese university students demonstrate the importance of examining sleep health in this population. Moreover, the link between SD and anxiety reminds us of the importance of treating anxiety as soon as detected and not simply targeting the reduction of sleep problems.


Introduction
The American National Commission of Sleep Disorders Research (NCSDR) defines insomnia as "a chronic or acute sleep disorder characterized by a complaint of difficulty initiating, and/ or maintaining sleep, and/or a subjective complaint of poor sleep quality that result in daytime impairment and subjective report of impairment" [1].
The different sets of questions and criteria used to assess and define insomnia result in a wide range of prevalence rates [2]. A general consensus has emerged from population-based studies. This consensus states that approximately 30% of a variety of adult samples from different countries have at least one of the symptoms of insomnia, such as difficulty initiating or maintaining sleep, waking up too early, and in some cases, having non-restorative or poor quality of sleep [3]. The DSM-IV-TR and research diagnostic criteria/International Classification of Sleep disorders, 2 nd Edition (RDC/ICSD-2) are commonly used diagnostic systems. These research tools have difficulties initiating or maintaining sleep in addition to daytime distress or impairment, but they differ in other symptoms required for diagnosis [2]. Recent studies have suggested that sleep dissatisfaction may be an important indicator of sleep pathology [4] and in the newly published DSM-5, dissatisfaction with sleep quality and quantity has been included as a necessary condition for insomnia diagnosis; in addition, the frequency of sleep problems for at least 3 nights per week and during a period of at least three months are required [5]. Insomnia is now recognized as a public health concern that affects the quality of life of millions of people around the world as it can lead to long-term physical and mental exhaustion with altered mood, concentration, and memory. Subsequently, the social and professional aspects of life are affected because of a deterioration of general condition with a decrease in intellectual abilities and cognitive behavior [6]. Sleep disorders (SDs) are particularly increasing in students as they face multiple stressors such as academic overload, constant pressure to succeed, and concerns about the future that alter the quality of their sleep. Sociodemographic factors such as age and gender, sleeping hygiene, physical illnesses and mental disorders were identified as the main correlates of insomnia [7]. Individuals with family histories of depression or anxiety and who manifest lifelong depression and anxiety beginning in childhood are at uniquely high risk for insomnia at midlife [8]. Moreover, a considerable number of researches documented a high prevalence of sleep problems among clinically-anxious youth [9,10] as well as elevated levels of anxiety and stress in university students [11]. A few studies suggest that anxiety may be a risk factor for future insomnia [12,13] and many suggest a bidirectional relationship between anxiety (and depression) and insomnia [14,15]. Among anxiety disorders, GAD (generalized anxiety disorder) has been identified as the disorder with the highest comorbidity of sleeping problems/insomnia [16].
Given the seriousness of insomnia and its repercussions on the academic and professional life of university students, the purpose of this study was to: 1) Assess SDs in students at the Campus of Medical Sciences (CMS) at Saint-Joseph University in Lebanon, as well as sociodemographic factors associated with insomnia, sleep quality, daytime sleepiness and anxiety; 2) Assess the relationship between different aspects of sleep (insomnia, quality of sleep and sleepiness) and anxiety. Our hypothesis were that SDs could be a major health concern in university students, and that examining anxiety using a simple tool (GAD-7) is primordial when exploring students sleep patterns.

Ethical considerations
The protocol of the study was approved by the ethics committee of Saint-Joseph University (Ref.USJ-2013-33, July 2013). Informed consent was obtained from all individuals participating in the study.

Survey procedure and sampling
Our study was a cross-sectional questionnaire-based survey conducted among students of three faculties: medicine, dentistry and pharmacy at Saint-Joseph University, from September 2013 to May 2014 (9 months). Inclusion criteria were: students aged 18 years and above, willing to participate in the study. Exclusion criteria were: age under 18 years and presence of chronic disease. Students were randomly selected within each class using a random number table to ensure the representativeness of the sample. This random selection was proportional to the number of students in each class. Students selected were approached by two trained research assistants usually at the end of their courses before leaving the classroom.

Data collection
Data were collected during a face-to-face interview using a self-administered standardized survey tool based on four internationally validated and reliable questionnaires, namely the Insomnia Severity Index [17], the Pittsburgh Sleep Quality Index (PSQI) [18], the Epworth Sleepiness Scale [19] and the Generalized Anxiety Disorder 7-item scale (GAD-7) [20]. The duration of interviews ranged from 10 to 20 minutes.
The PSQI is a 19-items questionnaire evaluating sleep quality and disturbances over the past month. The first four items are open questions, whereas items 5 to 19 are rated on a 4-point Likert scale. Individual items scores yield seven components (sleep disturbance, overall sleep quality, sleep latency, duration of sleep, daytime dysfunction due to sleepiness, sleep efficiency, and need for medicines to sleep). A total score, ranging from 0 to 21, was obtained by adding the seven component scores. Some studies stated that a score < 5 suggests a good sleep quality [23]. However, the vast majority of studies involving the PSQI state clearly that a score > 5 suggests poor sleep quality whereas a score 5 suggests a good sleep quality [18,24,25]. In our study, we chose the latest PSQI score interpretation.
The ESS is a self-administered questionnaire with eight questions. Each participant rated on a 4-point scale (0-3) his general level of daytime sleepiness, or the average sleep propensity in daily life. The total ESS score was the sum of eight item-scores and ranged between 0 and 24. The higher the score, the higher is the person's level of daytime sleepiness, with significant sleepiness when the score was > 10 [24,25]. Finally, GAD-7 is a 7-item instrument that assesses generalized anxiety severity. Each item was scored 0 to 3, providing a 0 to 21 severity score (0-4: normal; 5-9: mild anxiety; 10-14: moderate anxiety and 15-21: severe anxiety), with clinically significant anxiety detected when the total score was 10 or above [20,26].

Data analysis
The statistical analysis was carried out using SPSS software for Windows (version 19, Chicago, IL, USA). The significance level was set at 5%. Sample characteristics were summarized using the mean and the standard deviation (SD) for continuous variables and percentage for categorical variables. Insomnia prevalence rate was calculated using descriptive data, along with its corresponding 95% confidence interval (CI). For statistical comparison, analysis of variance or Student's t-tests were used for continuous variables, Chi-square test (χ 2 ) or Fisher Exact test were used for categorical variables and Spearman correlation coefficient test for correlations.

Socio-demographic characteristics of the participants
A total of 515 students were approached to participate in the study, of whom 462 (89.7%) consented. Our study population comprised 140 (31.3%) male and 322 (69.7%) female students. Age ranged between 18 and 30 years (mean 21.2±1.8 years). The sample included 154 students from the Faculty of medicine (FM), 108 from the Faculty of dentistry (FD) and 200 from the Faculty of pharmacy (FP).

Insomnia prevalence and severity (ISI)
The presence of insomnia was evaluated according to the ISI questionnaire. Prevalence of clinical insomnia was 10.6% with a 95% confidence interval ranging between 7.8 and 13.4%. The mean ISI score of the sample was 10.06±3.76. Table 1 displays the proportions of participants who endorsed each item response to the ISI questionnaire. Among 462 students, 129 (27.9%) had no insomnia (ISI 0-7), 284 (61.5%) had sub-threshold insomnia (ISI [8][9][10][11][12][13][14] and 49 (10.6%) had clinically significant insomnia (ISI > 14). Of these, 48 had moderate and only one had severe insomnia. Neither gender nor faculty was significantly associated with ISI score. The association between participants' age and ISI score did not reach statistical significance (p = 0.064); however, the year of study was found to be statistically significantly associated with the ISI score (p = 0.041), clinical insomnia being statistically significantly more frequent in first year students.

Subjective Sleep Quality (PSQI)
The mean PSQI score was 5.14±2.84 and 287 (62.9%) participants were good sleepers. Age, gender, faculty and year of study did not differ significantly between good and poor sleepers (p>0.05). However, sleep duration and sleep disturbance subscales differed significantly between the three faculties ( Table 2). Sleep duration was shortest for students of the FM while sleep disturbance was more pronounced for the FD students. Subjective sleep quality, sleep duration, sleep disturbance and daytime dysfunction were significantly associated with age and year of study. Only sleep efficiency subscale was significantly different between male and female participants (p = 0.039), worse for females.
The proportion of poor sleepers was higher in participants with clinical insomnia (p = 0.001); however, PSQI and ESS were not significantly associated (p>0.05).

Anxiety (GAD-7)
Anxiety was evaluated using the GAD-7 tool. The mean GAD-7 score was 7.01±4.81. Table 3 displays the proportion of participants who endorsed each item response to the GAD-7 questionnaire. Our study showed that 37.1% of the participants do not suffer from anxiety, 33.7% have mild anxiety, 21.6% moderate and 7.1% severe anxiety. Overall, 28.7% of students presented clinically significant anxiety. Anxiety was not significantly associated with any of the socio-demographic characteristics of the participants (p>0.05).

Daytime sleepiness (ESS)
Daytime sleepiness of the participants was evaluated using the ESS tool. The mean ESS score was 7.71±4.18. Of all participants, 23.9% scored above 10 on the ESS and were considered to have excessive daytime sleepiness. One item was associated with a high chance of dozing in nearly 30% of the students: "lying down to rest in the afternoon when circumstances permit" (Table 4). Daytime sleepiness was not significantly associated with any of the socio-demographic characteristics of the participants (p>0.05).
Excessive daytime sleepiness was significantly more frequent among participants with clinical insomnia (p = 0.031) than those with no clinical insomnia.
Associations between anxiety (GAD-7) and ISI, PSQI or ESS Clinically significant anxiety was significantly more frequent in students suffering from clinical insomnia (p = 0.006) and in poor sleepers (p = 0.003). Moreover, 50.8% of the participants with clinically significant anxiety presented excessive daytime sleepiness, versus 30.9% of those with no clinically significant anxiety (p<0.0001) ( Table 5).

Discussion
Insomnia is one of the most common SDs, with prevalence between 10% and 40% across studies [27,28]. Prevalence of sleep disorders among young adults is increasing globally and in adolescents aged 13-16, prevalence of insomnia was reported to be 11% [29].   Our study revealed that 10.6% of participants suffered from clinically significant insomnia. These results are consistent with the nature of the sample studied (young students) and is comparable with what is reported in young adults aged 20 to 29 (9.1%) [30] and in college students (12-13%) [31].
Quality of sleep is another important parameter to explore, since it was associated with quality of life in many studies [32,33]. Students classified as poor-sleepers experienced more problems with physical and psychological health. Improving sleep quality has been shown to decrease the incidence of chronic diseases such as major depressive disorders, psychosis, addictions, cardiovascular, metabolic and inflammatory disease risks [34]. Our study revealed that a considerable number (37.1%) of participants are poor sleepers and poor sleep quality was significantly related to the presence of clinical insomnia (p = 0.001).
Excessive daytime sleepiness is a symptom with high clinical and public health importance because of its association with increased risk for accidents, decreased productivity, metabolic syndrome, diabetes and impaired quality of life [35,36]. Insomnia symptoms and anxiety were associated with the subsequent occurrence of excessive daytime sleepiness in one study [35]. However, excessive daytime sleepiness was associated with depression, but not anxiety in another study [37]. Our results showed a significant association between anxiety and excessive daytime sleepiness: 50.8% of the participants with clinically significant anxiety presented excessive daytime sleepiness, versus 30.9% of the participants with no clinically significant anxiety Table 3. GAD-7 items with corresponding numbers (proportions) of participants endorsing each score (N = 436*).
Over the last 2 weeks, how often have you been bothered by the following problem?
Not at all Several days More than half the days Nearly everyday (p<0.0001). Furthermore, excessive daytime sleepiness was significantly more frequent in students with clinical insomnia (p = 0.031).This latest finding is of primary importance since previous publications assessing the relationship between insomnia and excessive daytime sleepiness show conflicting findings [35] [38]. Furthermore, our results showed no association between ESS and PSQI scores, which is a finding reported previously [18,25]. The explanation could be that these two questionnaires evaluate different aspects of sleep [24].
Overall, 28.7% of students presented clinically significant anxiety and our results are comparable to what is previously reported among university students [39].
One of the most robust associations between sleep disruption and emotional functioning involves the role of anxiety [40,41], and insomnia increases the likelihood of developing a mood or anxiety disorder [42]. In terms of public health, insomnia co-morbid with anxiety or depression gives rise to more complications and greater burdens than does each disorder alone [43] and treating anxiety improves co-morbid insomnia symptoms [42].
Our results showed that anxiety was significantly associated with insomnia as well as with sleep quality and daytime sleepiness.

Strength and limitations
Our findings should be interpreted in the context of the study's design and limitations. The results of our survey rely on self-reported behavior. Self-reporting questionnaires remain the most widely used tools in community surveys [44,45]. The self-report method reflects the interviewee's own perspective, which may be more suitable for reporting subjective disorders. The questionnaires were formulated in a "multiple-choice" and scale pattern to facilitate response and have shorter interview duration in order to avoid disturbing the students, in the hope that the simplicity of the questionnaire would make it easier for the respondents to give accurate informations. PSQI subscales show a significant difference between faculties and this probable bias needs to be highlighted. Sleep hygiene and sleep schedules have not been examined and need to be explored in future studies since they are important determinant of The numbers in bold represent statistically significant results.
behavioral insomnia in college and university students [31]. Furthermore, anxiety and not depression was assessed in our study. However, since the prevalence of anxiety in university students is much higher than depression [39] and since no causal relationship between anxiety and insomnia is proposed, we are confident that this limitation did not influence our results. Finally, comorbidities and use of drugs were not examined since the presence of any chronic disease was among the exclusion criteria of this study. Notwithstanding these limitations, the findings observed in this study are important and warrant further investigations.
To the best of our knowledge, this was the first study that examines three different aspects of sleep health in university students: insomnia, quality of sleep, and excessive daytime sleepiness. It is also the first that assesses the relationships between these three SDs and anxiety. ISI and PSQI were combined in numerous studies and some others combine them with ESS or to GAD-7 but none with both.
In conclusion, our study explored several aspects of sleep parameters in university students together with anxiety assessment. The magnitude of sleep perturbation, in this sample of university students demonstrates the importance of examining sleep health in this population. Anxiety is not only associated with insomnia but also with poor sleep quality and with excessive daytime sleepiness among these students. These findings underline the importance of anxiety evaluation when assessing sleep in similar populations. Due to the bidirectional pathways between anxiety and sleep, interventions aiming at breaking the circle of sleep-affecting anxiety and anxiety affecting sleep are needed. Anxiety management, along with other measures such as sleep-hygiene and stress management measures could improve students' sleep.