To develop and demonstrate interrater reliability for a Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) Sleep Disorders (SCISD).
The SCISD was designed to be a brief, reliable, and valid interview assessment of adult sleep disorders as defined by the DSM-5. A sample of 106 postdeployment active-duty military members seeking cognitive behavioral therapy for insomnia in a randomized clinical trial were assessed with the SCISD prior to treatment to determine eligibility. Audio recordings of these interviews were double-scored for interrater reliability.
The interview is 8 pages long, includes 20 to 51 questions, and takes 10 to 20 minutes to administer. Of the nine major disorders included in the SCISD, six had prevalence rates high enough (ie, n ≥ 5) to include in analyses. Cohen kappa coefficient (κ) was used to assess interrater reliability for insomnia, hypersomnolence, obstructive sleep apnea hypopnea (OSAH), circadian rhythm sleep-wake, nightmare, and restless legs syndrome disorders. There was excellent interrater reliability for insomnia (1.0) and restless legs syndrome (0.83); very good reliability for nightmare disorder (0.78) and OSAH (0.73); and good reliability for hypersomnolence (0.50) and circadian rhythm sleep-wake disorders (0.50).
The SCISD is a brief, structured clinical interview that is easy for clinicians to learn and use. The SCISD showed moderate to excellent interrater reliability for six of the major sleep disorders in the DSM-5 among active duty military seeking cognitive behavioral therapy for insomnia in a randomized clinical trial. Replication and extension studies are needed.
Taylor DJ, Wilkerson AK, Pruiksma KE, Williams JM, Ruggero CJ, Hale W, Mintz J, Organek KM, Nicholson KL, Litz BT, Young-McCaughan S, Dondanville KA, Borah EV, Brundige A, Peterson AL; on behalf of the STRONG STAR Consortium. Reliability of the structured clinical interview for DSM-5 sleep disorders module. J Clin Sleep Med. 2018;14(3):459–464.
Current Knowledge/Study Rationale: Structured clinical interviews are the gold standard method for reliably and accurately assessing mental disorders in clinical research and practice. This study developed and tested the reliability of the first Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) Sleep Disorders (SCISD).
Study Impact: A reliable and accurate assessment of sleep disorders is a crucial first step toward examining epidemiology and ensuring the best evidence-based interventions for disorders primarily diagnosed by self-report (eg, insomnia, restless legs syndrome, nightmare disorder). The SCISD produces reliable diagnoses for six DSM-5 sleep disorders, providing the type of measure essential for standardization and advancement of treatment efficacy trials and clinical epidemiology.
Structured clinical interviews are the gold standard for assessing mental disorders in clinical research.1,2 They are superior to typical nonstructured clinical interviews at deriving reliable and accurate diagnoses for psychiatric2 and sleep disorders.3 A clinical interview is essential for diagnosing many sleep-wake disorders (eg, insomnia disorder, restless legs syndrome, nightmare disorder, circadian rhythm sleep-wake disorders).4,5 Self-report measures alone are insufficient because they are unable to accurately capture the multifaceted nature of sleep, which requires clinical judgment to assess. Also, these disorders cannot be assessed with an overnight sleep study because they are based on patterns of sleep over the course of multiple nights. Other sleep disorders, such as obstructive sleep apnea hypopnea (OSAH) and rapid eye movement (REM) sleep behavior disorder, require an overnight sleep study for diagnosis because an objective assessment of problematic occurrences during sleep (eg, apnea, etc.) is required for a diagnosis.6
A structured clinical interview for sleep disorders is needed to ensure reliable and accurate diagnoses to guide evidence-based treatment selection and to help standardize and advance treatment research and clinical epidemiology in the field of behavioral sleep medicine.7 Several structured or semistructured clinical interviews are currently available to assess sleep disorders.3,8–10 To date, these instruments either have not been tested8,10 or do not assess for Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5)11 defined sleep disorders.3,8–10
The Structured Interview for Sleep Disorders (SIS-D) is an early published structured interview that was based on the DSM, Third Edition, Revised (DSM-III-R).9 This measure was designed for “experienced psychiatric interviewers” (ie, 2 years beyond final medical degree) and requires 20 to 30 minutes to administer. Interrater reliability was assessed in 68 patients referred to a sleep laboratory for complaints of sleep disturbances. These authors found “excellent” interrater reliability for insomnia (0.91), hypersomnia (0.90), sleep-wake schedule disorders (0.79), dream anxiety (0.79), and sleep walking (0.79) disorders.
However, the DSM has undergone 2 substantial revisions in the interim, so this assessment instrument is not current.
Around the same time that the SIS-D was released, the Insomnia Interview Schedule (IIS) was developed.10 It was designed to facilitate the collection of a detailed sleep history according to the DSM, Fourth Edition (DSM-IV),12 the International Classification of Sleep Disorders (ICSD),13 and commonly used research criteria10 of the time. The IIS includes 95 to 100 questions and requires about 20 to 30 minutes to administer by a clinician or mental health professional with some knowledge of sleep disorders. It gathers a wide variety of information, including nature and severity of the insomnia complaint, sleep/wake schedule, use of sleep medications and aids, symptoms of other sleep disorders, medical and psychiatric history, and current or past psychopathology. The most notable limitation of the IIS is that it was primarily built for the assessment and diagnosis of insomnia and so does not adequately assess for other disorders (eg, circadian rhythm sleep-wake disorders-shift work type, nightmare disorder). Further, the reliability and validity of the IIS has not been established.
Edinger et al.8 developed the Duke Structured Interview Schedule (Duke SIS) with the release of the DSM, Fourth Edition-Text Revision (DSM-IV-TR)14 and the second edition of the ICSD (ICSD-2).15 The format of the Duke SIS was also updated to be consistent with the widely used Structured Clinical Interview for DSM-IV Disorders,16,17 which assesses a wide array of mental health diagnoses. The Duke SIS includes approximately 150 questions and requires 30 to 90 minutes to administer, depending on how many sleep problems the patient endorses. The primary limitation of the Duke SIS is the extensive length. The Duke SIS also focuses primarily on categorizing the many different insomnia types listed in ICSD-2 (eg, adjustment insomnia, psychophysiological insomnia, idiopathic insomnia, inadequate sleep hygiene) and does not perform in-depth assessments of other sleep disorders. Similar to the IIS, reliability and validity of the Duke SIS was never established, and it, too, is based on outdated diagnostic criteria.
Recently, Merikangas et al.3 developed and validated a Diagnostic Interview for Sleep Patterns and Disorders (DISP), which is also based on the ICSD-2 and requires 30 minutes to administer. The study compared the referring clinician diagnosis with a blind evaluation with DISP in 225 patients.3 The comparisons were established with “…information on the clinical history, clinical examination, nocturnal [polysomnography], and [Multiple Sleep Latency Test], when relevant.” Using this study design, which is different from traditional interrater reliability studies, the authors found “excellent” reliability for narcolepsy with cataplexy (0.79), and “acceptable” reliability for delayed sleep phase (0.54), hypersomnia (0.41), and periodic limb movement disorder (0.47). “Poor” reliability was found for insomnia (0.29), OSAH (0.38), and restless legs syndrome (0.20). Thus, the DISP is limited by insufficient evaluation of the interrater reliability and by the focus on the ICSD-2, which is only used by sleep specialists and not other providers who see patients with sleep disorders (eg, psychologists, social workers, primary care physicians, etc.).
The Structured Clinical Interview for DSM-5 Sleep Disorders (SCISD; sleep.unt.edu/SCISD) was developed to assist in the efficient and accurate assessment of sleep disorders in two large studies that began recruitment during the transition from the DSM-IV-TR to DSM-5.18,19 The first author (DJT), a licensed psychologist certified in behavioral sleep medicine (CBSM) and two graduate students (AKW, KMO) used the existing criteria of the DSM-IV-TR in conjunction with the proposed changes for the DSM-5 to create individual interview questions for all the sleep-wake disorder categories. Of the proposed changes, only those that were ultimately included in the DSM-5 are in the final version of the SCISD. The SCISD fully assesses for six sleep-wake disorders and screens for three sleep-wake disorders that require polysomnography for confirmation. The SCISD was designed to allow an independent evaluator with a master's degree or higher and minimal training to obtain a reliable sleep disorder diagnosis in adults. It requires 10 to 20 minutes to administer, which is shorter than the interviews described previously. A more detailed description of the measure is included in the Measures section of this manuscript.
The purpose of the current study was to determine the inter-rater reliability of the SCISD using data from a randomized clinical trial comparing in-person and Internet cognitive behavioral therapy of insomnia to a minimal contact control.19
Prior to enrolling in the parent study,19 participants completed a phone screen to ensure subjective complaints about sleep and stability on medications. Exclusion criteria included current suicidal intent, pregnancy, or serious mental health diagnosis such as bipolar disorder or history of psychosis. Those who met initial screening criteria were scheduled for a baseline appointment, where they completed a variety of self-report questionnaires and then were administered the SCISD by an independent evaluator. More detailed description of the procedures, including operational definitions and detailed inclusion and exclusion criteria, can be found in the parent study.19
Assessments were administered by master's and doctoral level independent evaluators with primary background and training in psychology with a focus on trauma and posttraumatic stress disorder. Evaluators had no prior training in sleep medicine. Evaluators completed a 3-hour training led by the first author, who is a licensed clinical psychologist and CBSM. Training included assigned readings, didactic presentations on the basics of sleep, DSM-5 sleep-wake disorders, and the administration and scoring of the SCISD. Evaluators also completed two to three mock interviews and one to two supervised interviews with patients prior to completing independent assessments. Further, evaluators completed weekly, ongoing supervision with the CBSM provider or psychologist with specialized training in behavioral sleep medicine. Eight independent evaluators were trained for the current study. Each month, at least 1 co-rating was completed with a randomly selected audio recording of a SCISD to assess interrater reliability.
Audio recordings of 106 SCISD assessments (92 baseline, 14 post-waitlist controls) were randomly selected for co-rating by one of the eight evaluators. The trained independent evaluators reviewed the recording and scored each module as described in the next paragraphs. Evaluators did not complete co-ratings for their own assessments.
The sample consisted of 106 participants randomly selected from 409 participants who provided informed consent and who completed a baseline assessment as part of the parent trial.19 Participants were recruited from April 2012 to December 2014 and were active-duty military personnel with at least one deployment in support of Operation Enduring Freedom, Operation Iraqi Freedom, or Operation New Dawn, and who were referred for treatment of sleep disturbances. The study was approved by the Institutional Review Boards at Brooke Army Medical Center, the University of Texas Health Science Center at San Antonio, and the University of North Texas as well as the United States Army Medical Research and Materiel Command Human Research Protection Office.
Structured Clinical Interview for DSM-5 Sleep Disorders (SCISD)
The final version of the SCISD fully assesses for insomnia disorder, hypersomnolence disorder, circadian rhythm sleep-wake disorders, two of the parasomnia disorders (non-REM sleep arousal disorder [ie, sleep walking and sleep terrors] and nightmare disorder), and restless legs syndrome. It further provides an initial screening of REM sleep behavior disorder, narcolepsy, and OSAH to help determine if a referral for diagnostic polysomnography is indicated. The other sleep-related breathing disorders (ie, central sleep apnea and sleep-related hypoventilation) were not included, as the symptoms needed for differentiation are not as readily apparent to the individual or the individual's bed partner and are subsequently difficult to identify in an interview format. Additionally, “other specified” and “unspecified” sleep-wake disorders were not included, as they lack specific criteria necessary to create diagnostic interview questions.
The SCISD is 8 pages long, with 20 to 51 questions (many can be potentially skipped due to rule-outs). Eight open questions related to typical schedule (eg, bedtime, sleep onset latency, rise time, etc.) begin the interview to assist in obtaining pertinent diagnostic information (eg, self-imposed sleep deprivation due to family and work demands). Subsequent questions assess every diagnostic symptom of the aforementioned disorders and are rated as 1 (absence of symptom), 2 (possible presence of symptom), 3 (definite presence of symptom), or “?” (unsure). If the necessary DSM-5 symptoms are endorsed for a given diagnosis, the participant is classified as having the disorder.
Data Analysis Plan
All variables were screened for errors and missingness. Inter-rater reliability between the original assessment and the co-rated assessment was evaluated using Cohen kappa (κ).20 Those disorders that did not have sufficient prevalence rates (ie, < 4%) were not included. This required at least five individuals to receive a diagnosis from at least one rater to be considered for analysis.9,21 Consistent with recommendations for use with the DSM-5,21 κ coefficients benchmarks were as follows: below 0.2 were poor, 0.2–0.4 were acceptable, 0.4–0.6 were good, 0.6–0.8 were very good, and greater than 0.8 were excellent.
The sample characteristics are presented in Table 1. The sample was aged 18 to 65 years (mean ± standard deviation: 33.0 ± 7.8), with 11% female. Rater 1 (primary rater) made the diagnosis for the following proportions of the sample: insomnia disorder = 93%, hypersomnolence disorder = 8%, circadian rhythm sleep-wake disorder = 8%, non-REM sleep arousal disorder = 2%, nightmare disorder = 30%, restless legs syndrome = 9%, REM sleep behavior disorder = 3%, narcolepsy = zero, and OSAH = 53%. Because non-REM sleep arousal disorder, REM sleep behavior disorder, and narcolepsy were < 4%, they were not included in subsequent analyses.
Participant characteristics (n = 106).
Participant characteristics (n = 106).
As seen in Table 2, interrater reliability analyses found insomnia (κ = 1.0) and restless leg syndrome (κ = 0.83) were excellent.21 Reliability for nightmare disorder (κ = 0.78) and OSAH (κ = 0.73) were very good. Hypersomnolence (κ = 0.50) and circadian rhythm sleep-wake disorder (κ = 0.50) had good reliability (all Ps < .001).
Interrater reliability of DSM-5 diagnoses based on SCISD administrations (n = 106).
Interrater reliability of DSM-5 diagnoses based on SCISD administrations (n = 106).
These findings provide preliminary interrater reliability evidence for the SCISD in a sample of active-duty military seeking cognitive behavioral therapy of insomnia within a randomized clinical trial.19 The SCISD had excellent interrater reliability for insomnia and restless legs syndrome, which are disorders largely based on self-report of symptoms. The SCISD also had very good reliability for nightmare disorder and sleep apnea. Finally, good reliability was found for hypersomnolence disorder and circadian rhythm sleep-wake disorder.
The SCISD reliability for insomnia (1.0) was better than the SIS-D (0.91), which used interrater reliability procedures similar to the current study. The difference in results may have been related to the nearly doubled sample size of insomnia diagnoses in the current study (99 versus 54), as the participants in the current study were referred for insomnia treatment rather than for any type of sleep disorder as in the SIS-D study. Similarly, the SCISD reliability for nightmare disorder (0.78) was roughly equivalent to the SIS-D category “dream anxiety disorder” (0.79). The current study still had a nearly sixfold greater sample of nightmare disorder diagnoses (29 versus 5). Our reliability for restless legs syndrome (0.83) was almost equivalent to those in the SIS-D study (0.85), which are high, considering both studies had very small sample sizes of this disorder (8 and 4).
The SCISD had very good reliability for OSAH (0.73), which was modestly higher than reliability found in the SIS-D9 for “sleep-induced respiratory impairment” (0.64). The current study had over 9 times more apnea cases than the SIS-D study (56 versus 6), which may have accounted for the higher reliability values.21 It is important to note that there is symptom overlap between OSAH symptoms (ie, middle-of-the-night awakenings and excessive daytime sleepiness) and the disorders of insomnia and hypersomnia. This symptom overlap may have inflated the rates of all three disorders, but as this was not a prevalence study, the overlap should not be a major concern as to biasing our reliability estimates.
The SCISD had good reliability (0.50) for circadian rhythm sleep-wake disorders, compared to excellent reliability (0.79) for the SIS-D9 on “sleep-wake schedule disorders.” Similar results were found for hypersomnolence disorder, where the SCISD performed worse (0.50) than the SIS-D (0.90). Given the relatively small sample sizes of this overall disorder category across these two studies (circadian rhythm sleep-wake disorder n = 4–7; hypersomnolence n = 7–14), it is difficult to interpret significant meaning from this difference. As suggested for the overall DSM-5,21 larger field trials are needed to investigate the true reliability of the SCISD assessment of these disorders.21
Due to insufficient sample size (ie, < 4%) the current study and the SIS-D study did not report on non-REM arousal disorder, REM behavior sleep disorder, or narcolepsy. The current study also did not report on periodic limb movement disorder because it is no longer a category in the DSM-5.
The major strengths of this study are the large sample size and the use of two independent, blind evaluators. The SCISD takes only 10 to 20 minutes to administer all modules and showed good to excellent interrater reliability for insomnia disorder, restless legs syndrome, nightmare disorder, OSAH, hypersomnolence, and the overall circadian rhythm sleep-wake disorder category. Therefore, the SCISD should be strongly considered for clinical trials or epidemiological studies where accurate and reliable assessment for these disorders is required. Of note, the questions used in the OSAH component of the interview were derived from a combination of the STOP sleep apnea questionnaire22 and a question asking participants if they received a previous diagnosis of OSAH with overnight polysomnography, which may have improved reliability.
The primary limitation of the current study was generalizability of the sample. The participants were primarily male (89%) active-duty patients seeking treatment for insomnia as part of a randomized clinical trial and who had passed a phone screen for symptoms of insomnia in order to participate. The sex ratio was representative of the United States Army (ie, 86% male).23 The high percentage of males might influence the rates of disorders identified by the structured clinical interviews, because research has shown that females are more likely than males to report psychological and physiological symptoms.24 Interestingly, that study did not find sex differences for insomnia, nor did our own research in a similar military sample.25 However, there is no research indicating reliability of structured clinical diagnoses would be expected to be different in women versus men. This may be an area for future research, but for now it can be assumed that the results reported here are generalizable to other randomized clinical trials of insomnia interventions, especially in active-duty military.
In addition, the study was limited by the relatively small proportion of the sample in whom hypersomnolence, circadian rhythm disorders, or REM sleep behavior disorder were diagnosed, and the inadequate proportion (ie, < 4%) in whom non-REM sleep arousal disorder and narcolepsy were diagnosed. However, previous studies of patients referred from sleep disorders centers had similar limitations.
Another limitation of the current study is that validity data were not provided. Previous studies, such as the DISP study, have attempted to provide validity of the various diagnoses using “expert” clinical interviews with access to clinical history, examination, and polysomnography when indicated. This “valid” referent is not without criticism. For instance, in the DISP study, the expert clinical interview found fewer “secondary” sleep disorders, particularly insomnia and restless legs syndrome, presumably because the clinician was only focused on the “primary” disorder. However, contemporary science indicates that disorders such as insomnia, restless legs syndrome, and sleep apnea are actually more accurately deemed “comorbid” rather than “secondary,” and this definition is now indicated in the DSM-5. In addition, studies of expert clinical interview diagnoses of psychiatric disorders routinely show that the structured interview is more accurate and reliable than self-report assessments, making it unclear which approach should be used as the reference category.26 Finally, even objective polysomnography shows night-to-night variability on accuracy of diagnosis of more organic sleep disorders such as obstructive apnea and hypopnea, with studies showing that as many as 43% to 50% of adult patients not meeting criteria on 1 night of polysomnography, subsequently meeting criteria on a second night.27,28
A further limitation was that reliability estimates were based on audio recordings, rather than live interviews performed by two independent clinicians. To a degree, that concern is mitigated by the use of a structured interview with standardized primary questions.29 Nevertheless, reliability estimates would likely have been smaller with two independent interviewers, not only because of potentially different delivery and follow-up questions, but because of changes in the respondent from one occasion to another. However, confidence in the current results is strengthened by their similarity to those of previous studies.
Future studies need to be performed in larger, more diverse samples with adequate representation in each of the DSM-5 sleep disorders. This may be accomplished by using a variety of settings, including sleep disorders centers, primary care settings, and the general population. Finally, some degree of validity assessment with overnight sleep studies, expert interviewers, etc., should be attempted to further bolster confidence in the results.
Work for this study was conducted at Carl R. Darnall Army Medical Center, Fort Hood, TX. All authors have seen and approved this manuscript. The authors report no conflicts of interest. Funding for this work was made possible by the US Department of Defense through the US Army Medical Research and Materiel Command, Congressionally Directed Medical Research Programs, Psychological Health and Traumatic Brain Injury Research Program award W81XWH-10-1-0828 (PI: Taylor). The views expressed in this article are solely those of the authors and do not reflect an endorsement by or the official policy or position of the United States Army, the Department of Defense, the Department of Veterans Affairs, or the United States Government.
certified in behavioral sleep medicine
Diagnostic Interview for Sleep Patterns and Disorders
Diagnostic and Statistical Manual of Mental Disorders Duke
Duke Structured Interview Schedule
International Classification of Sleep Disorders
Insomnia Interview Schedule
obstructive sleep apnea hypopnea
rapid eye movement
Structured Clinical Interview for DSM-5 Sleep Disorders
Structured Interview for Sleep Disorders
The authors thank: Lucas Brilliott, MS, Monica Gauna, MS, Dana Larson, MA, Susan Paschall, MS, Thomas Vaccaro, PhD, and Veronica Vargus, LMSW, University of Texas Health Science Center at San Antonio, who supported this study as independent evaluators and co-raters; Jamie Anderson, BS, Michelle Barrera, BS, Julia Garza, BA, Amanda Hasselle, BA, Nicholas Holder, BS, Sonia Holloman, BS, University of Texas Health Science Center at San Antonio, and Jessica Dietch, MS, and Jade Francetich, MS, University of North Texas, who supported this study as study coordinators and research assistants; Susan Deason, BA, Gary Burk, MA, MBA, and Deanne Hargita, MPA, University of Texas Health Science Center at San Antonio, who served as part of the STRONG STAR Administrative Core; Raymond Aguilar, BS, Ivan Negovetic, PhD, and Kevin Muenzler, BS, University of Texas Health Science Center at San Antonio, who served as part of the STRONG STAR Data and Biostatistics Core; Crystal Mendoza, MA, University of Texas Health Science Center at San Antonio, who supported this study as part of the STRONG STAR Assessment Core; Steffany Malach, PhD, University of Texas Health Science Center at San Antonio, who helped develop the fidelity rating procedures; Kimberly Del Carmen, PhD and Holly Campbell-Rosen, PhD, Department of Defense, Congressionally Directed Medical Research Programs, who supported this study as grant officer representatives; and Julie Collins, BA, and Joel Williams, BA, BS, University of Texas Health Science Center at San Antonio, who provided editorial support for this manuscript.
Rogers R. Handbook of Diagnostic and Structured Interviewing. New York, NY: Guilford Press; 2001.
APA Work Group on Psychiatric Evaluation. Practice Guideline for Psychiatric Evaluation of Adults. Arlington, VA: American Psychiatric Association; 2016.
Merikangas K, Zhang J, Emsellem H, et al. The structured Diagnostic Interview for Sleep Patterns and Disorders: rationale and initial evaluation. Sleep Med. 2014;15(5):530–535. [PubMed]
Chesson A, Hartse K, McDowell W, et al. Practice parameters for the evaluation of chronic insomnia. Sleep. 2000;23(2):237–242. [PubMed]
Morgenthaler T, Lee-Chiong T, Alessi C, et al. Practice parameters for the clinical evaluation and treatment of circadian rhythm sleep disorders: an American Academy of Sleep Medicine report. Sleep. 2007;30(11):1445–1459. [PubMed Central][PubMed]
Chesson A, Ferber RA, Fry JM, et al. Practice parameters for the indications for polysomnography and related procedures. Sleep. 1997;20(6):406–422. [PubMed]
Buysse DJ, Ancoli-Israel S, Edinger JD, Lichstein KL, Morin CM. Recommendations for a standard research assessment of insomnia. Sleep. 2006;29(9):1155–1173. [PubMed]
Edinger JD, Kirby AC, Lineberger MD, et al. Duke Structured Interview Schedule for DSM-IV-TR and International Classification of Sleep Disorders, Second Edition, Sleep Disorders Diagnoses. Durham, NC: Veterans Affairs and Duke University Medical Centers; 2006.
Schramm E, Hohagen F, Grasshoff U, et al. Test-retest reliability and validity of the Structured Interview for Sleep Disorders according to DSM-III-R. Am J Psychiatry. 1993;150(6):867–872. [PubMed]
Morin CM. Insomnia: Psychological Assessment and Management. New York, NY: The Guilford Press; 1993.
American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Arlington, VA: American Psychiatric Association; 2013.
American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: American Psychiatric Association; 1994.
American Sleep Disorders Association. The International Classification of Sleep Disorders, Revised : Diagnostic and Coding Manual. Rochester, MN: American Sleep Disorders Association; 1997.
American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. text revision. Washington, DC: American Psychiatric Association; 2000.
American Academy of Sleep Medicine. International Classification of Sleep Disorders: Diagnostic and Coding Manual. 2nd ed. Westchester, IL: American Academy of Sleep Medicine; 2005.
First MB, Spitzer RL, Gibbon M, Williams JBW. Structured Clinical Interview for DSM-IV Axis I Disorders, Clinician Version (SCID-CV). Washington, DC: American Psychiatric Press, Inc.; 1996
First MB, Spitzer RL, Gibbon M, Williams JBW. Structured Clinical Interview for DSM-IV Personality Disorders, (SCID-II). Washington, DC: American Psychiatric Press, Inc.; 1997.
Taylor DJ, Kelly K, Kohut ML, Song KS. Is insomnia a risk factor for decreased influenza vaccine response? Behav Sleep Med. 2017;15(4):270–287. [PubMed]
Taylor DJ, Peterson AL, Pruiksma KE, et al; the STONG STAR Consortium. Internet and in-person cognitive behavioral therapy for insomnia in military personnel: a randomized clinical trial. Sleep. 2017;40(6). [PubMed]
Cohen J. A coefficient of agreement for nominal scales. Educ Psychol Meas. 1960;20:37–46
Kraemer HC, Kupfer DJ, Clarke DE, Narrow WE, Regier DA. DSM-5: how reliable is reliable enough? Am J Psychiat. 2012;169(1):13–15. [PubMed]
Chung F, Yegneswaran B, Liao P, et al. STOP questionnaire: a tool to screen patients for obstructive sleep apnea. Anesthesiology. 2008;108(5):812–821. [PubMed]
United States of America Department of Defense. 2015 Demographics: Profile of the Military Community. Washington, DC: United States of America Department of Defense; 2015.
Kroenke K, Spitzer RL. Gender differences in the reporting of physical and somatoform symptoms. Psychosom Med. 1998;60(2):150–155. [PubMed]
Taylor DJ, Pruiksma KE, Hale WJ, et al; for the STRONG STAR Consortium. Prevalence, correlates, and predictors of insomnia in the US Army prior to deployment. Sleep. 2016;39(10):1795–1806. [PubMed Central][PubMed]
Basco RM, Bostic JQ, Davies D, et al. Methods to improve diagnostic accuracy in a community mental health setting. Am J Psychiatry. 2000;157(10):1599–1605. [PubMed]
Meyer TJ, Eveloff SE, Kline LR, Millman RP. One negative polysomnogram does not exclude obstructive sleep apnea. Chest. 1993;103(3):756–760. [PubMed]
Mosko SS, Dickel MJ, Ashurst J. Night-to-night variability in sleep apnea and sleep-related periodic leg movements in the elderly. Sleep. 1988;11(4):340–348. [PubMed]
Conway JM, Jako RA, Goodman DF. A meta-analysis of interrater and internal consistency reliability of selection interviews. J Appl Psychol. 1995;80(5):565–579