ADVERTISEMENT

Issue Navigator

Volume 12 No. 06
Earn CME
Accepted Papers
Classifieds







Scientific Investigations

Psychometric Comparison of the Full and Abbreviated Versions of the Dysfunctional Beliefs and Attitudes about Sleep Scale

http://dx.doi.org/10.5664/jcsm.5878

Ka-Fai Chung, MBBS, MRCPsych1; Fiona Yan-Yee Ho, MPhil2; Wing-Fai Yeung, PhD3
1Department of Psychiatry, The University of Hong Kong, Hong Kong SAR, China; 2Department of Psychology, The University of Hong Kong, Hong Kong SAR, China; 3School of Chinese Medicine, The University of Hong Kong, Hong Kong SAR, China

ABSTRACT

Study Objectives:

The different versions of the Dysfunctional Beliefs and Attitudes about Sleep Scale (DBAS) have limited comparison and summary of the findings across studies. We aimed to examine which version and which subscales had better psychometric properties.

Methods:

Data were derived from a randomized controlled trial of internet-based cognitive-behavioral therapy for insomnia (CBT-I) vs. waitlist in 312 participants with self-report diagnosis of insomnia disorder. The response distribution, internal consistency, construct and concurrent validity, and sensitivity to change were analyzed.

Results:

Floor or ceiling effects were found in 19 of the 30 DBAS items. Item-total correlation was < 0.30 in 43.3%, 31.3%, and 10.0% of the items in DBAS-30, DBAS-16, DBAS-10. Internal consistency was satisfactory for total scores, with Cronbach α ranging from 0.73–0.81, but 2 subscales of DBAS-30 and 1 subscale of DBAS-10 had Cronbach α < 0.35. Factor analysis produced 8, 4, and 3 factors for DBAS-30, DBAS-16, and DBAS-10. Only the factor structure of DBAS-16 was compatible with previous studies. Concurrent validity with insomnia, anxiety, and depressive symptoms was much stronger than with sleep diary parameters. Sensitivities to change of the DBAS scores following CBT-I and with sleep improvement were found, except the DBAS-30 “attributions” subscale and DBAS-16 “medication” subscale.

Conclusions:

The DBAS-16 possesses better internal consistency, a reproducible factor structure, strong concurrent validity, and sensitivity to change, and therefore is recommended for research use. The DBAS-30 and DBAS-10 have their own strengths, but there are limitations in their application as a quantitative measure in research.

Citation:

Chung KF, Ho FY, Yeung WF. Psychometric comparison of the full and abbreviated versions of the dysfunctional beliefs and attitudes about sleep scale. J Clin Sleep Med 2016;12(6):821–828.


INTRODUCTION

The Dysfunctional Beliefs and Attitudes about Sleep Scale (DBAS), which first appeared in Morin's insomnia treatment manual as a pre-treatment evaluation tool, has now become one of the most commonly used scales for the assessment of various sleep-related cognitions.1 Studies have shown that people with insomnia have higher DBAS scores than good sleepers. As an outcome measure, several of the DBAS items are sensitive to cognitive-behavioral therapy for insomnia (CBT-I) and their changes correlate with sleep improvement.2 There are different versions of the DBAS, but the most commonly used are the 30, 16, and 10 item versions. The 30-item DBAS covers 5 themes, including (1) consequences of insomnia, (2) control and predictability of sleep, (3) sleep requirement expectations, (4) causal attributions of insomnia, and (5) sleep promoting practices. However, studies have shown that only the subscales on “consequences” and “control and predictability” achieve satisfactory internal consistency. In addition, principal component analysis failed to achieve item convergence.3 For convenience of use and better psychometric properties, abbreviated versions of the DBAS have been developed. The 10-item version was created based on the items which had significant pre-post changes following CBT-I,3 while the 16-item version was based on response distribution, missing rate, item-total correlation, and lack of overlap with other items.4 Preliminary data suggested that DBAS-10 and DBAS-16 were valid and reliable.3,4 However, Carney and Edinger5 found that only 2 of the 30 DBAS items possessed the ability to differentiate insomniacs from good sleepers, were sensitive to change following CBT-I, and correlated with sleep improvement, and the authors remained open to which DBAS version should be used. The DBAS has been translated into several languages, including Chinese. Chen et al.6 in Taiwan showed that only 2 of the 5 subscales of the 30-item version and 2 of the 4 subscales of the 16-item version had satisfactory internal consistency, and the factor structure was not supported by confirmatory factor analyses. To better understand the psychometric properties of the DBAS, we examined the 30, 16, 10 items versions in 312 Chinese subjects with insomnia disorder. The research questions were to find out the strengths and weaknesses of each version and whether the DBAS subscales were valid and reliable.

BRIEF SUMMARY

Current Knowledge/Study Rationale: The different versions of the Dysfunctional Beliefs and Attitudes about Sleep Scale (DBAS) have limited comparison and summary of the findings across studies. We aimed to examine which version and which subscales had better psychometric properties.

Study Impact: The DBAS-16 possesses better internal consistency, a reproducible factor structure, strong concurrent validity, and sensitivity to change. Future studies should consider the DBAS-16 as a better option to quantify sleep-related cognitions in research settings.

METHODS

Sample

Data of the present study were derived from a randomized controlled trial of self-help CBT-I.7 Adults aged 18 years or above with insomnia disorder compatible with the Diagnostic and Statistical Manual, Fifth Edition (DSM-5) criteria,8 including difficulty initiating or maintaining sleep, early morning awakening, or non-restorative sleep associated with significant distress or impairment in occupational, social and other important areas of functioning ≥ 3 nights per week for > 3 months, participated in our study through the internet. No monetary incentive was provided, but the treatment was free of charge. The only exclusion criterion was report of current suicidal ideation.

Procedure

All procedures used in the present study were reviewed and approved by the local institutional review board. Subjects who showed an interest in the study were directed to a website. Informed consent was obtained prior to all study procedures. After completing a questionnaire on sociodemographic and clinical variables, eligible subjects were provided with a user-name and password. Three-hundred twelve participants were randomly assigned to self-help CBT-I with telephone support (n = 103), self-help CBT-I (n = 104), and waiting-list control (n = 105).

Intervention

CBT-I was provided over 6 weeks, with treatment materials delivered once per week. The content was adopted from a well-established CBT-I manual.9 We summarized the most salient information of the treatment manual and edited the materials to be relevant in the local setting. Most of the information was presented in text, together with some diagrams, and a 15-minute audio clip on relaxation training. Details of the program are available.7 In brief, session 1 consists of a program overview, basic facts about sleep, etiological model of insomnia, and goal setting. Session 2 includes sleep hygiene, activity scheduling, and relaxation training. Session 3 begins with a revision of the sleep hygiene rules, followed by an introduction of stimulus control and sleep restriction. Session 4 begins with a revision of the stimulus control and sleep restriction principles, followed by cognitive therapy. Session 5 focuses on cognitive therapy and relaxation training. The last session summarizes the major points of previous sessions, reviews the treatment goal, and covers relapse prevention.

In the group allocated to receive CBT-I with telephone support, a psychology graduate (YYH) contacted the participants weekly for roughly 15 min each time to answer questions about treatment content, review treatment progress, and encourage the participants to read the materials and complete the assignments. We found that the self-help CBT-I with or without telephone support was more effective than waitlist for improving sleep at both immediate and 4-week post-treatment.7

Measures

DBAS

A 30-item DBAS on a 10-point Likert scale was used.10 Participants indicated their degree of endorsement of the items with “strongly disagree” at the far left and “strongly agree” at the far right. Higher scores represent stronger endorsement of the beliefs. We edited the language and sentence structure of the Taiwanese version6 to form the Hong Kong Chinese version.

Expanded Consensus Sleep Diary for Morning (CSD-M)

The CSD-M is a prospective self-report measure developed by a group of experts in an effort to standardize sleep diary assessment. Participants were asked to complete the CSD-M everyday within 1 hour of getting out of bed for 7 consecutive days.11 The sleep parameters were averaged over 1 week. Only total sleep time and sleep efficiency were analyzed in order to avoid multiple comparisons.

Insomnia Severity Index (ISI)

The ISI is a 7-item self-rating scale that was used to assess the perceived severity of insomnia and the associated functional impairment in the previous week.12 The items were rated using a 5-point Likert scale with anchor points ranging from “none” to “very severe,” “very satisfied” to “very dissatisfied,” and “not at all noticeable, worried, or interfering” to “very much noticeable, worried, or interfering.” The total score ranges from 0 to 28. We used the Chinese version of ISI, known to be valid and reliable.13

Hospital Anxiety and Depression Scale (HADS)

The HADS is a self-report scale that was used to assess the severity of anxiety and depression in the past week.14 It comprises 2 subscales, each with 7 items, measuring the severity of anxiety and depressive symptoms. The items use different anchor descriptors and are scored on a 4-point scale. The sub-scale score ranges from 0 to 21, with higher scores indicating more severe symptoms. The HADS was chosen because it was well validated in the Chinese population.15

Data Analysis

All statistical analyses were performed using SPSS version 23 for Windows (SPSS, Chicago, USA). Response distribution and floor and ceiling effects were revealed using the mean, SD, and % with lowest and highest score. The contribution of each item to the total score was assessed using item-total correlation. Internal consistency was evaluated with Cronbach α: as its interpretation depends on the number of scale items,16 the mean and range of inter-item correlations were presented. Factor structure was examined by principal component analysis. If factor correlations exceed 0.32, which is equivalent to more than 10% overlap in variance, promax rotation would be used.17 The factors were selected according to eigenvalues > 1 and the Scree test. The items with rotated loadings ≥ 0.50 were regarded as representing significant item-factor contributions. With a sample size of 312 in this study, the subject-to-item ratio was 10.4, which was well above the minimum of 5 as recommended in textbooks on factor analysis.18 Concurrent validity was assessed by correlating the DBAS scores with clinical parameters that are likely to be associated with sleep-related cognitions, including sleep diary measures, insomnia, depressive, and anxiety symptoms. The baseline data were used to analyze response distribution, internal consistency, factor structure, and concurrent validity. Sensitivity to change was assessed by comparing the DBAS change scores from baseline to immediate post-treatment between CBT-I and waitlist group and between responders and non-responders. In line with a previous study,19 participants with ISI scores improved by ≥ 8 points were classified as responders. There are different interpretations of the validity and reliability indices; in this study, significant floor or ceiling effect was defined as > 20% of respondents using the lowest or highest score.20 Corrected item-total correlations ≥ 0.30 were considered acceptable.21 The coefficients for internal consistency and concurrent validity were regarded as satisfactory if they were between 0.70 and 0.95 and between 0.30 and 0.70, respectively.22

RESULTS

Sample Characteristics

Table 1 presents the demographics and clinical characteristics of the participants. Participants had a mean age of 38.5 years and 71.1% were female. The mean ISI score was 15.7, the mean sleep efficiency over 7 days was 78.5% (range = 37.7% to 96.7%), and the mean sleep onset latency and wake after sleep onset were both around 47 minutes.

Sample characteristics.

jcsm.12.6.821.t01.jpg

table icon
Table 1

Sample characteristics.

(more ...)

Descriptive Statistics, Item-Total Correlation, and Internal Consistency

Significant floor effect was found in items 3, 11, 13, 14, 15, 16, 26, 27, and 28. Significant ceiling effect happened in 10 items (Table 2). The floor and ceiling effects were most severe in 5 items (items 5, 14, 15, 26, and 27), of which greater than 40% of participants used the extreme scores. There were 13 items (43.3%) of the DBAS-30 having item-total correlation < 0.30; for DBAS-16, items 1, 11, 19, 24, and 27 had similar problems (31.3% of total), but only 1 item in DBAS-10 (10.0%) had an item-total correlation < 0.30. The internal consistency of the DBAS-30, DBAS-16, and DBAS-10 were satisfactory, with Cronbach α 0.81, 0.80, and 0.73, respectively (Table 3). Several subscales had Cronbach α > 0.70, including DBAS-30 “consequences” and “control and predictability” subscales, DBAS-16 “consequences” and “worry/helplessness” subscales, and DBAS-10 “immediate consequences” subscale, while some subscales, including DBAS-30 “sleep expectations,” “attributions,” and “sleep promoting practices” subscales and DBAS-10 “control and predictability” subscale had poor internal consistency (< 0.40) and low mean inter-item correlation (< 0.20).

Descriptive statistics, internal consistency and item-total correlation of the DBAS.

jcsm.12.6.821.t02.jpg

table icon
Table 2

Descriptive statistics, internal consistency and item-total correlation of the DBAS.

(more ...)

Internal consistency of the DBAS-30, DBAS-16, and DBAS-10 and their subscales.

jcsm.12.6.821.t03.jpg

table icon
Table 3

Internal consistency of the DBAS-30, DBAS-16, and DBAS-10 and their subscales.

(more ...)

Factor Analysis

Principal component analysis with promax rotation of the DBAS-30 items found 8 factors with eigenvalue > 1 and the same number of factors was indicated by the Scree plot. The 8 factors explained 55.1% of the rotated variance (Table 4). Factor 1 consisted of items 10, 12, 17, 18, 21, and 30 and resembled the “consequences” subscale, of which 5 of the 9 subscale items were included. Factor 3 consisted of items 1, 2, 6, 7, and 9, which had certain similarities with the “sleep promoting practices” sub-scale, of which 4 of the 7 subscale items were included. Factors 5 consisted of items 3, 13, and 16, which was similar to the “sleep expectations” subscale. However, Factors 2, 4, and 6–8 were not compatible with any of the DBAS-30 subscales. Several items had rotated loadings marginally above or below the criterion of 0.50, suggesting that the factor structure was not ideal and some items might be relevant to more than 1 factor. For DBAS-16, principal component analysis produced 4 factors, which explained 55.5% of the rotated variance. Factor 1 consisted of items 10, 12, 17, 18, 21, and 30, which was similar to the “consequences” subscale. Factor 2 included items 5, 8, 19, 20, and 25, which resembled the “worry/helplessness” subscale. Factor 3 had the same items as the “sleep expectations” subscale. Factor 4 consisted of items 11 and 27, which was similar to the “medication” subscale. Item 24 was not loaded on any of the 4 factors, while items 5, 10, 19, 20, and 25 had rotated loadings marginally above or below 0.50, indicating their relevance to more than 1 factor. Principal component analysis of the DBAS-10 items produced 3 factors, which accounted for 55.6% of the variance. Factor 1 consisted of items 10, 12, 17, and 21, which included 3 of the 5 items of the “immediate consequences” subscale. Factor 2 consisted of items 1, 2, and 7, which had limited resemblance with the DBAS-10 subscales. Factor 3 included items 5 and 8 and resembled “long-term consequences” subscale. Item 22 was not loaded on any of the 3 factors.

Principal component analysis (promax rotation) of the DBAS-30, DBAS-16, and DBAS-10.

jcsm.12.6.821.t04.jpg

table icon
Table 4

Principal component analysis (promax rotation) of the DBAS-30, DBAS-16, and DBAS-10.

(more ...)

Concurrent Validity

There were significant correlations between DBAS total and several subscale scores with ISI and PSQI total scores and HADS-anxiety and HADS-depression subscale scores. However, the correlations with sleep diary-derived total sleep time and sleep efficiency were mostly not significant. The DBAS-30 “attributions” and “sleep promoting practices” subscales were not correlated with any of the insomnia, anxiety, and depressive symptom scales, while the DBAS-30 and DBAS-16 “sleep expectation” subscales were only significantly correlated with ISI total score. The DBAS-10 “immediate consequences” sub-scale had significant correlation with ISI and HADS-anxiety subscale, but was not related to HADS-depression subscale.

Sensitivity to Change

Significantly greater reduction in DBAS scores was found in participants allocated to CBT-I than those on waitlist, except the DBAS-30 “attributions” subscale and DBAS-16 “medication” subscale. Participants who had improved in ISI scores by ≥ 8 points had significantly greater changes in DBAS total scores, DBAS-30 “control and predictability” and “sleep promoting practices” subscale scores, DBAS-16 “worry/helpfulness” subscale score, and DBAS-10 “long-term consequences” and “control and predictability” subscale scores, compared to those who had not improved. A descriptive summary of the psychometric properties of DBAS-30, DBAS-16, and DBAS-10 is shown in Table 5.

Descriptive summary on the psychometric properties of DBAS-30, DBAS-16, and DBAS-10

jcsm.12.6.821.t05.jpg

table icon
Table 5

Descriptive summary on the psychometric properties of DBAS-30, DBAS-16, and DBAS-10

(more ...)

DISCUSSION

Despite the frequent use of the DBAS as a measure of sleep-related cognitions, we found shortcomings in the psychometric properties of its full and abbreviated versions. Although the full version has a broad coverage and may be suitable for pre-treatment evaluation, it has unstable factor structure, many of its items have floor or ceiling effects, and several of its sub-scales have unsatisfactory internal consistency; hence its use as a research instrument can be problematic. The DBAS-16 has better overall psychometric properties, including better internal consistency, a reproducible factor structure, strong concurrent validity, and sensitivity to change. The DBAS-10 was created based on the items which had significant pre-post changes following CBT-I; hence it may have a restricted indication. As one the subscales of the DBAS-10 has unsatisfactory internal consistency, instead of analyzing its subscales, the DBAS-10 should be used as a global measure.

Floor or ceiling effects were found in 19 of the 30 items of DBAS, probably because the DBAS was initially designed to assess dysfunctional cognitions; hence some of the items were worded to identify the most severe kinds of concern. For example, the items “fear of dying in sleep,” “insomnia seriously affects health,” and “medication as the only solution” may produce affirmative negative responses. We replicated the previous findings in Western and Chinese populations that several DBAS subscales had unsatisfactory internal consistency. Two subscales of the DBAS-30 and 1 subscale of the DBAS-10 had the worst internal consistency, ranging from 0.01 to 0.33.

Exploratory factor analysis of the DBAS-30 produced 8 factors, which had limited resemblance to its 5 subscales, probably because the subscales were originally constructed based on clinical intuition than on psychometric properties. The DBAS-16 had the most consistent factor structure among the 3 versions tested. We replicated the 4-factor structure of the DBAS-16 found in a Caucasian population with insomnia.4 A difference was observed in item 17 “one poor night disturbs whole week,” which was loaded on the “consequences” factor in our population, but not the “worry/helplessness” factor in the Caucasian sample; in addition, item 24 “insomnia resulting from chemical imbalance” was not loaded on any of the factors in our sample, perhaps due to a lack of agreement among participants in their perceived causes of insomnia23 and its independence from other sleep-related cognitions. In addition, several items belonging to the “consequences” and “worry/ helplessness” factors had rotated loadings closed to the criterion of 0.50, suggesting that the two constructs may have some overlap in meaning. For DBAS-10, our factor analysis was not entirely compatible with two previous studies,3,24 between which differences in factor structure were also detected.

We found that all three versions of DBAS had acceptable concurrent validity. Similar to the finding in a previous study,4 the correlation of DBAS with sleep diary parameters was weaker than with insomnia, anxiety, and depressive symptom scales, suggesting that quantitative measures of insomnia played a limited role in dysfunctional sleep-related cognitions. We showed that the “attributions” and “sleep promoting practices” subscales of the DBAS-30 were not correlated with insomnia, anxiety, and depressive symptoms. The findings were compatible with previous studies that the scores on these 2 subscales were similar between insomniacs and good sleepers.5,25 Although some of the items in these 2 subscales could be regarded as dysfunctional cognitions, e.g., “need to catch up on sleep loss,” “more time in bed insures more sleep,” “trying harder will lead to sleep,” these beliefs may be cultural values held by the general public and are not specific to people with insomnia.

The DBAS was found to be sensitive to changes. Except the “attributions” subscale of the DBAS-30 and “medication” sub-scale of the DBAS-16, all other subscales either improved after CBT-I or changed in line with sleep improvement. The “consequences” and “sleep expectations” subscales of the DBAS-16 improved after CBT-I but not in line with sleep improvement. With better understanding of the psychometric properties of various versions of the DBAS and hopefully a consensus as to which version to use, comparison and summary of the findings from different studies can be made. Further studies on the role of DBAS subscales as mediators of sleep improvement are needed.

There are several methodological limitations in the study. A requirement to have access to the internet and basic computer skills may have limited the participants to better educated individuals. The diagnosis of insomnia disorder may not be accurate as it is based on self-report, without confirmation by clinical interview. Another limitation was that the DBAS-16 and DBAS-10 were not completed on their own, but as part of the DBAS-30, hence we did not know whether item rating was affected. Lastly, we did not include good sleepers to test the discriminant validity of the DBAS in a Chinese population nor conducted test-retest reliability assessment.

In conclusion, the DBAS-16 is a valuable quantitative instrument for the assessment and monitoring of sleep-related cognitions in people with insomnia. The DBAS-10 and DBAS-30 have relatively weak factor structure and internal consistency, hence the use of their subscales for research purpose is not recommended. Our findings suggest that the DBAS-10 should be used as a global measure, preferably for treatment outcome studies. Additional studies are needed to further validate the DBAS-16 subscales as a classifier of insomnia subtype, an outcome predictor, and for tailoring treatment.26

DISCLOSURE STATEMENT

This was not an industry supported study. The authors have indicated no financial conflicts of interest.

ABBREVIATIONS

CBT-I

cognitive-behavioral therapy for insomnia

CSD-M

Consensus Sleep Diary for Morning

DBAS

Dysfunctional Beliefs and Attitudes about Sleep Scale

DSM

Diagnostic and Statistical Manual of Mental Disorders

HADS

Hospital Anxiety and Depression Scale

ISI

Insomnia Severity Index

ACKNOWLEDGMENTS

Author contributions: Dr. K.F. Chung designed the study, interpreted the results, and prepared the manuscript. Ms. Y.Y. Ho and Dr. W.F. Yeung collected and analyzed data. We declare that all authors have substantial contributions to the study, including drafting or revising the work.

REFERENCES

1 

Morin CM, editor. Insomnia: psychological assessment and management. New York, NY: Guilford Press, 1993.

2 

Schwartz DR, Carney CE, authors. Mediators of cognitive-behavioral therapy for insomnia: a review of randomized controlled trials and secondary analysis studies. Clin Psychol Rev. 2012;32:664–75. [PubMed]

3 

Espie CA, Inglis SJ, Harvey L, Tessier S, authors. Insomniacs' attributions: psychometric properties of the Dysfunctional Beliefs and Attitudes about Sleep Scale and the Sleep Disturbance Questionnaire. J Psychosom Res. 2000;48:141–8. [PubMed]

4 

Morin CM, Vallieres A, Ivers H, authors. Dysfunctional Beliefs and Attitudes about Sleep (DBAS): validation of a brief version (DBAS-16). Sleep. 2007;30:1547–54. [PubMed Central][PubMed]

5 

Carney CE, Edinger JD, authors. Identifying critical beliefs about sleep in primary insomnia. Sleep. 2006;29:444–53. [PubMed]

6 

Chen CW, Jan YW, Yang CM, Lin SC, authors. Dysfunctional Beliefs and Attitudes about Sleep (DBAS): validation of the Chinese version. Arch Clin Psychol. 2009;4:59–67. [in Chinese].

7 

Ho FY, Chung KF, Yeung WF, Ng TH, Cheng SK, authors. Weekly brief phone support in self-help cognitive behavioral therapy for insomnia disorder: relevance to adherence and efficacy. Behav Res Ther. 2014;63C:147–56. [PubMed]

8 

American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 5th ed. Arlington, VA: American Psychiatric Publishing, 2013.

9 

Morin CM, Espie CA, editors. Insomnia: a clinician's guide to assessment and treatment. New York, NY: Springer, 2003.

10 

Morin CM, Stone J, Trinkle D, Mercer J, Remsberg S, authors. Dysfunctional beliefs and attitudes about sleep among older adults with and without insomnia complaints. Psychol Aging. 1993;8:463–7. [PubMed]

11 

Carney CE, Buysse DJ, Ancoli-Israel S, et al., authors. The consensus sleep diary: standardizing prospective sleep self-monitoring. Sleep. 2012;35:287–302. [PubMed Central][PubMed]

12 

Bastien CH, Vallières A, Morin CM, authors. Validation of the Insomnia Severity Index as an outcome measure for insomnia research. Sleep Med. 2001;2:297–307. [PubMed]

13 

Chung KF, Kan KK, Yeung WF, authors. Assessing insomnia in adolescents: comparison of Insomnia Severity Index, Athens Insomnia Scale and Sleep Quality Index. Sleep Med. 2011;12:463–70. [PubMed]

14 

Zigmond AS, Snaith RP, authors. The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand. 1983;67:361–70. [PubMed]

15 

Leung CM, Wing YK, Kwong PK, Lo A, Shum K, authors. Validation of the Chinese-Cantonese version of the Hospital Anxiety and Depression Scale and comparison with the Hamilton Rating Scale of Depression. Acta Psychiatr Scand. 1999;100:456–61. [PubMed]

16 

Cortina JM, author. What is coefficient alpha? An examination of theory and applications. J Appl Psychol. 1993;78:98–104.

17 

Gaskin CJ, Happell B, authors. On exploratory factor analysis: a review of recent evidence, an assessment of current practice, and recommendations for future use. Int J Nurs Stud. 2014;51:511–21. [PubMed]

18 

Watson R, Thompson DR, authors. Use of factor analysis in Journal of Advanced Nursing: literature review. J Adv Nur. 2006;55:330–41.

19 

Morin CM, Vallières A, Guay B, et al., authors. Cognitive behavioral therapy, singly and combined with medication, for persistent insomnia: a randomized controlled trial. JAMA. 2009;301:2005–25. [PubMed Central][PubMed]

20 

Mao HF, Msueh IP, Tand PF, Sheu CF, Hsieh CL, authors. Analysis and comparison of the psychometric properties of three balance measures for stroke patients. Stroke. 2002;33:1022–7. [PubMed]

21 

Nunnally J, Bernstein I, editors. Psychometric theory. New York, NY: McGraw-Hill, 1994.

22 

Streiner DL, author. A checklist for evaluating the usefulness of rating scales. Can J Psychiatry. 1993;38:140–8. [PubMed]

23 

Yung KP, Chung KF, Ho FY, Yeung WF, Ng TH, authors. The experience of chronic insomnia in Chinese adults: a study using focus groups and insomnia experience diaries. Behav Sleep Med. 2015 Sep 22. [Epub ahead of print].

24 

Edinger JD, Wohlgemuth WK, authors. Psychometric comparisons of the standard and abbreviated DBAS-10 versions of the dysfunctional beliefs and attitudes about sleep questionnaire. Sleep Med. 2001;2:493–500. [PubMed]

25 

Edinger JD, Fins AI, Glenn DM, et al., authors. Insomnia and the eye of the beholder: are there clinical markers of objective sleep disturbances among adults with and without insomnia complaints? J Consult Clin Psychol. 2000;68:586–93. [PubMed]

26 

Sánchez-Ortuño MM, Edinger JD, authors. A penny for your thoughts: patterns of sleep-related beliefs, insomnia symptoms and treatment outcome. Behav Res Ther. 2010;48:125–33. [PubMed Central][PubMed]