INTRODUCTION
Insomnia is a pervasive sleep disorder that affects a significant portion of the population worldwide [1,2]. Insomnia is characterized by difficulty initiating or maintaining sleep, resulting in inadequate sleep duration. The impact of insomnia extends beyond nighttime discomfort, as it affects daily functioning, physical and mental health, and overall well-being. Numerous studies have investigated the causes, consequences, and treatment approaches for insomnia. Understanding the underlying factors contributing to insomnia is important to improve sleep outcomes, general health, and quality of life. Identifying the specific comorbidities associated with insomnia is essential for comprehending the complex interplay between sleep and other physical and mental health conditions.
Rating scales, such as the Insomnia Severity Index (ISI) [3], are popularly used to assess insomnia severity in clinical practice. Assessing insomnia using scales provides a comprehensive evaluation based on the scale chosen. We aimed to develop an insomnia severity assessment tool that clinicians could naturally incorporate during patient interviews in clinical practice. We sought to devise a method to indirectly ascertain the severity of insomnia using essential parameters such as the sleep index that must be invariably conducted during insomnia evaluations. Consequently, we proposed a measure of the discrepancy between the desired time in bed and the desired total sleep time, namely the DBST index [4]. The DBST index assesses the mismatch between an individual’s desired time in bed (dTIB) and their desired total sleep time (dTST). The DBST is considered a contributing factor to insomnia and aims to comprehensively evaluate insomnia severity. Several studies have investigated DBST as a relevant factor in investigating sleep problems across various populations and clinical conditions.
According to the concept paper, among the general population [4], insomnia severity, along with depression, preoccupation with sleep, and dysfunctional beliefs about sleep, was associated with the DBST index. Mediation analysis revealed that depression, preoccupation with sleep, and dysfunctional beliefs about sleep mediated the influence of the DBST index on insomnia severity. In the general population [5], insomnia severity was associated with depression, preoccupation with sleep, dysfunctional beliefs about sleep, and the DBST index. Furthermore, preoccupation with sleep mediated the influence of the DBST index on individuals’ insomnia severity. These findings indicate that the DBST index may replace the ISI for measuring insomnia severity. Further, greater discrepancies between the dTIB and dTST are associated with increased insomnia severity among patients with cancer [6]. In addition, the DBST index was one of the expected variables for sleep onset latency (SOL).
We considered that the DBST index could be incorporated into Cognitive-Behavioral Therapy for Insomnia (CBT-I). We explored whether changes in the DBST index corresponded to changes in insomnia severity among the general population; a reassessment was conducted in the same group. The results indicated a significant correlation between the change in the DBST index and the change in the ISI, suggesting that changes in the DBST index are associated with changes in insomnia severity [7]. The results address the possibility of using the DBST index as an aid for CBT-I for patients with insomnia.
The association between the DBST index and insomnia severity has not been explored among clinical samples of patients with insomnia. Previous studies have reported a relationship among the general population; however, evidence of this relationship among patients with insomnia, whose insomnia severity may be high, is lacking. Furthermore, patients with insomnia may experience depression or anxiety simultaneously [8]. Clinically, insomnia can appear as a primary symptom; however, it can occur secondary to depression or anxiety [9]. Thus, the relationship between the DBST index and insomnia severity needs to be explored among clinical samples, considering the effect of depression or anxiety. We posited that 1) the DBST index may be positively associated with insomnia severity, 2) the influence of the DBST index on insomnia severity may be mediated by depression, and 3) anxiety symptoms may mediate the influence of the DBST index on insomnia severity.
METHODS
Participants and procedure
We conducted a retrospective medical record review of patients with insomnia who visited the Sleep Clinic in Asan Medical Center, Seoul, Korea. We reviewed the medical records of all patients with insomnia who visited the clinic for the first time from May 2021 to March 2022. Patients who 1) could not walk, 2) had organic problems with impaired cognitive function, 3) were diagnosed with major psychosis, 4) could not complete the rating scales, or 5) had a problem with communication were excluded. Data on age, sex, comorbid sleep, and psychiatric disorder diagnoses of 127 patients were collected from medical records. In addition, their sleep indices and responses to self-rating scales were collected. The study protocol was approved by Asan Medical Center’s Institutional Review Board (2022-0517). Obtaining written informed conset was waived by IRB.
Sleep indices
At the first clinic visit, the clinician routinely asked the patients questions such as “What is your usual bedtime?”, “What is your usual time to fall asleep?”, “What is your usual time to finally get out of bed in the morning?”, “For how many hours do you want to sleep in a day?”, and “From when until when do you want to sleep?”. We estimated the patients’ DBST index, bedtime, sleep onset time, wake-up times, SOL, time spent in bed (TIB), duration from wake-up time to bedtime (WTB), and duration of time spent in bed during 24 h (TIB/d) from the responses to the questions [10].
Time variables
Time variables (bedtime, sleep onset time, and wake-up time) were estimated by averaging the usual times reported by patients. For example, if a patient answered that they usually go to bed between 23:00 and 24:00, the usual bedtime was estimated at 23:30. The usual times were transformed into numerical variables for statistical analysis: 15 min into 0.25, and 30 min into 0.50. Hence, 23:30 was transformed into 11.5 [11].
Duration variables
Duration variables (SOL, TIB, WTB, or TIB/d) were obtained using the estimated time variables. For example, SOL was estimated from the duration between bedtime and sleep onset time, and the TIB was estimated between bedtime and wake-up time. The WTB was estimated from the duration between wake-up time and bedtime (i.e., WTB=24-TIB) [10].
Discrepancy between desired TIB and desired TST: DBST index
The DBST index was estimated from the difference between the patient’s dTST and TIB [4]. The clinician asked the patient, “How many hours do you want to sleep a day? (dTST)” and “From what time to what time do you want to sleep? (dTIB).” The dTST was estimated by averaging the responses. For example, for a patient who responded, “I want to sleep for 7 or 8 h,” we estimated the dTST to be 7.5. The dTIB was estimated from the time the participants reported. For example, if a patient responded, “I want to sleep from 11:00 pm to 7:00 am,” we calculated the dTIB to be 8 h. The DBST index was calculated as the desired hours of TIB-desired hours of TST [4].
Rating scales
Insomnia Severity Index
The ISI is a self-rating scale for assessing insomnia severity [3]. It comprises seven items measured on a 5-point Likert-like scale, resulting in a total score ranging from 0 to 28. A higher total score indicates more severe insomnia. We applied the validated Korean version of the ISI in this study [12], and the Cronbach’s alpha for our sample was 0.841.
Patient Health Questionnaire-9
The Patients Health Questionnaire-9 (PHQ-9) is a self-reported scale developed to assess an individual’s depression severity [13]. The PHQ-9 comprises nine items, each rated on a 4-point Likert scale ranging from 0 (not at all) to 3 (nearly every day); the total PHQ-9 score ranges from 0 to 27. A higher total score on the PHQ-9 reflects a more severe level of depression (0–4=minimal depression, 5–9=mild depression, 10–14=moderate depression, 15–19=moderately severe depression, and ≥20=severe depression). The Cronbach’s alpha for the sample was 0.856, indicating strong internal consistency.
Generalized Anxiety Disorder-7
The Generalized Anxiety Disorder-7 (GAD-7) is a self-administered questionnaire designed to evaluate general anxiety levels [14]. It comprises seven items, each rated on a 4-point Likert scale (0=not at all to 3=nearly every day), with total scores ranging from 0 to 21. Higher scores indicate more pronounced generalized anxiety symptoms. The questionnaire, developed initially by Spitzer et al. [14], recommends a cutoff score of 10 to identify generalized anxiety disorder cases, with anxiety categorized as minimal (0–4), mild (5–9), moderate (10–14), or severe (15–21). The Cronbach’s alpha for this sample was 0.899.
Dysfunctional Beliefs and Attitudes about Sleep-16 items
The Dysfunctional Beliefs and Attitudes about Sleep-16 items (DBAS-16) is a self-reported rating scale designed to assess individuals’ dysfunctional beliefs and attitudes regarding sleep [15]. The 16 items of the DBAS-16 are rated on a Likert-type scale from 0 (strongly disagree) to 10 (strongly agree), with the final score calculated by averaging the scores for all 16 items. A higher total score signifies a greater extent of dysfunctional beliefs about sleep. This study utilized the Korean version of the DBAS-16 scale [16]. The Cronbach’s alpha for the sample was 0.890.
Epworth Sleepiness Scale
The Epworth Sleepiness Scale (ESS) was developed to measure possible excessive daytime sleepiness [17]. Individuals self-evaluate their likelihood of falling asleep in each situation using a 4-point scale (0=not at all likely to doze; 1=slight likelihood of dozing; 2=moderate likelihood of dozing; 3=high likelihood of dozing). A total score exceeding 10 indicates excessive daytime sleepiness symptoms. The Cronbach’s alpha for this sample was 0.852.
Statistical analyses
The demographic variables and rating scale scores are summarized as the mean±standard deviation. A two-tailed level of significance at p<0.05 was used. Student’s t-test was utilized for continuous variables, while the chi-squared test was employed for categorical variables. Pearson’s correlation analysis was conducted for correlation assessments. Multiple linear regression analysis was performed to investigate clinical variables contributing to the ISI score and the DBST index, controlling for age and the PHQ-9, GAD-7, DBAS-16, and ESS scores. Additionally, a causal mediation analysis was conducted to explore whether the relationship between the DBST index and insomnia severity may be mediated by the PHQ-9, GAD-7, DBAS-16, or ESS scores. The statistical analyses were conducted using SPSS version 21.0 (IBM Corp., Armonk, NY, USA) and AMOS version 27 for Windows (IBM SPSS., Chicago, IL, USA) and RStudio (Posit, Boston, MA, USA).
RESULTS
A total of 127 patients were included in the analyses (Table 1). Their mean age was 59.3±12.3 years. Most patients (n=111, 87.4%) were diagnosed with insomnia disorder. The others were diagnosed with depressive disorders (8.7%), anxiety disorders (3.1%), adjustment disorders (0.8%). Regarding sleep indices, the means for the time and duration variables are shown in Table 2.
Table 3 describes the results of the correlation analyses. Patient age was correlated with ISI (r=-0.19, p<0.05), PHQ-9 (r=-0.20, p<0.05), dTST (r=-0.22, p<0.05), and DBST (r=0.21, p<0.05). The ISI scores were significantly correlated with the PHQ-9 (r=0.42, p<0.01), GAD-7 (r=0.34, p<0.01), DBAS-16 (r=0.51, p<0.001), and DBST (r=0.20, p<0.05) indices. DBST index scores were significantly correlated with age (r=0.21, p<0.05), ISI scores (r=0.20, p<0.05), dTST (r=-0.52, p<0.001), and dTIB (r=0.32, p<0.01).
A multiple linear regression analysis was conducted (Table 4) to explore the association of different variables with the ISI score, including age, DBST, PHQ-9, GAD-7, DBAS-16, ESS, and DBST index scores. Age (β=-0.18, p=0.018), DBST (β=0.23, p=0.003), PHQ-9 (β=0.23, p=0.031), and DBAS-16 (β=0.42, p<0.001) contributed to insomnia severity (adjusted R2=0.37, F=13.1, p<0.001). In contrast, the DBST index was affected by age (β=0.26, p=0.005) and the ISI score (β=0.33, p=0.003).
Table 5 and Figure 1 show the results of causal mediation analysis conducted using R software. The mediation hypotheses were rejected since the results did not show significant indirect effects of depression, anxiety, dysfunctional beliefs about sleep, or daytime sleepiness on the association between the DBST index and insomnia severity. Regardless of mediators, DBST directly influenced insomnia severity.
DISCUSSION
In this study, we explored whether the DBST index could be a significant factor for insomnia severity among patients with insomnia. We observed that the DBST index was associated with depression and dysfunctional beliefs about sleep that contributed to insomnia severity in clinical samples. Furthermore, the DBST index directly influenced insomnia severity regardless of the mediating effects of depression, anxiety, dysfunctional beliefs about sleep, or daytime sleepiness.
The DBST index has contributed the ISI score among the general population [5]. In this study, we observed similar results among clinical samples of patients with insomnia. A general population study with a wide range of ISI scores showed a significant correlation between DBST and ISI scores. This research determined whether the DBST index remains significant in a group in which the severity of insomnia is relatively greater than that in the general population, such as patients with insomnia, in which the range of ISI scores may be comparatively narrower. We observed that the DBST index remained a significant predictor of the ISI even among the insomnia group, suggesting that it can contribute to the ISI when controlling for the PHQ-9, GAD-7, DBAS-16, and ESS.
The DBST index was not significantly correlated with depression, anxiety, dysfunctional beliefs about sleep, or daytime sleepiness in this study. Similar results have been observed among patients with cancer, who may have more severe insomnia [18]. These results show that the DBST index can be a tool for measuring insomnia severity rather than other components that can influence sleep quality, such as depression or dysfunctional beliefs about sleep. These results confirm that the DBST index can be used solely to assess insomnia severity, even in clinical samples of patients with insomnia.
However, a relationship between the DBST index and ISI has not been consistently observed. We previously reported no correlation between the DBST index and ISI in a group of shift-working nursing professionals [19]. Possible explanations for the lack of association included the potential contribution of higher dTST, as participants consciously sought more sleep due to demanding work and the influence of younger age in participants engaged in shift work. Moreover, the DBST index may have been estimated during nonshift work periods while shift working.
This study, however, did not find indirect effects of psychological factors, such as depression, daytime sleepiness, dysfunctional beliefs about sleep, or anxiety, on the relationship between DBST and ISI. Correlation analyses revealed that the DBST index correlated only with ISI scores without showing any significant correlation with PHQ-9, GAD-7, DBAS-16, ESS, or other psychological elements. In the general population, these factors were associated with DBST; however, this association was not observed among patients with insomnia. A total of 111 (87.4%) of the participants had primary insomnia, and their average ISI score was 16.9±5.8. In contrast, depression- and anxiety-related scores were not markedly elevated. As a result, other psychological factors may not have significantly influenced the impact of the DBST index on ISI, possibly due to the relatively low depression and anxiety levels observed within this specific group of patients. However, a significant correlation was observed between the ISI score and depression level in this study. The findings indicate that the DBST is an effective tool for assessing insomnia severity, excluding depression and anxiety, which are readily observed among patients with insomnia. This discrepancy needs to be examined in further studies.
This study had several limitations. First, it was conducted in a single tertiary referral hospital. Patients with severe insomnia may visit the clinic, and the characteristics of the patients enrolled in this study should be considered when interpreting the results. Second, objective measurements, such as polysomnography, were not performed. Objective tests need to be conducted in subsequent studies to explore the DBST index and objective measurements among patients with insomnia disorders. In addition, because of the lack of polysomnography data, other sleep disorders, including obstructive sleep apnea and periodic limb movement disorder, could not be ruled out. Third, the effects of medications could not be adjusted for in this study. Patients’ pharmacological or nonpharmacological treatment might impact their total sleep time, insomnia severity, or daytime sleepiness. This study aimed to explore the association between the DBST index and insomnia severity among a clinical sample of insomnia patients who might have undergone various insomnia treatments. Nevertheless, patients’ pharmacological and nonpharmacological treatment histories should be explored and considered in subsequent studies. Fourth, there was a negative correlation between age and insomnia severity in this study. These findings suggest that insomnia was more severe among the younger group than in the elderly group. However, younger patients who visit the sleep clinic may suffer more severe insomnia symptoms since they cannot easily visit the clinic due to work. Furthermore, some studies have reported a negative correlation between age and insomnia severity [20,21].
In conclusion, this study showed that the DBST index could be a tool for evaluating insomnia severity in a clinical sample of patients with insomnia, regardless of dysfunctional beliefs about sleep, anxiety, depression, or daytime sleepiness. The direct effect of DBST on insomnia severity highlights its independent and pivotal role in influencing sleep outcomes among patients with insomnia disorder.
