Semi-Structured Clinical Interviews

One commonly used interview is the Structured Clinical Interview for Sleep Disorders-Revised (SCISD-R). This is designed for trained health professionals familiar with the Diagnostic and Statistical Manual V (DSM-5) to screen for sleep disorders. It covers insomnia, hypersomnolence, circadian rhythm sleep-wake, nightmare, rapid eye movement sleep arousal, and rapid eye movement sleep behavior disorders, as well as OSA, restless legs syndrome (RLS), and narcolepsy [18, 19]. It is useful for standardization in research studies to identify presence of possible sleep disorders, as well as for diagnostic clarification in behavioral sleep medicine settings. The SCISD-R can be especially beneficial when questionnaire data is inconclusive, or when sleep complaints are complex. The SCISD-R is the only widely used semi-structured clinical interview dedicated exclusively to sleep disorders, though other notable mentions include the sleep module of the Mini International Neuropsychiatric Interview [20] and the sleep disorders section of the Structured Clinical Interview for DSM-5 [18].

Subjective Sleep Measures

Subjective sleep assessment includes self-report prospective sleep diaries and retrospective questionnaires. Sleep diaries and questionnaires are among the most cost-effective sleep assessments [21]. Most sleep disorders (OSA being a notable exception) can be diagnosed based on patient report. Thus, subjective sleep measures play a critical role in tracking symptomatology in both clinical and research settings. This section will describe the use of sleep diaries and the most frequently used questionnaires in research for measuring sleep quality, chronotype, and sleep disorder symptomatology. Clinically oriented measures are included for the consideration of researchers given that the goal of some studies may require identification and recruitment of participants with clinical diagnoses or selection of a clinically important patient outcome measure in a behavioral or pharmacological intervention clinical trial study (e.g., ensuring the successful treatment of insomnia in order to examine subsequent benefits on cardiovascular health metrics). Both subjective and objective sleep measures are summarized in Table 1.

Table 1 Summary of included subjective and objective sleep measuresSleep Diaries

Sleep diaries are short, post-sleep questionnaires that are completed for at least seven consecutive days [22]. Typical questions include estimates for: time spent in bed, number of hours slept, time it took to fall asleep, number of nighttime awakenings, wake time, and activities performed the previous day that may have impacted sleep (e.g., medication, alcohol, caffeine, exercise) [22]. From sleep diaries, it is possible to calculate the following components of the sleep window: SOL, WASO, total sleep time (TST), total time in bed (TIB), and SE (which is a percentage calculated by dividing TST by TIB and then multiplying by 100). Most sleep diaries also have a Likert rating of sleep quality or satisfaction [23]. Sleep diaries are available in both physical and electronic (web- and application-based) formats, with electronic sleep diaries having the capacity to automate scoring, document the day and time a sleep diary is completed, and help avoid “parking lot syndrome,” which is when a person retrospectively fills out their diary with multiple nights of sleep at once [24].

The Consensus Sleep Diary was created to help facilitate consistency of the data being collected from sleep diaries in both clinical and research settings [23]. There are nine items on the core Consensus Sleep Diary: (1) time the respondent got into bed; (2) time the respondent attempted to fall asleep; (3) SOL; (4) number of awakenings; (5) duration of awakenings; (6) time the respondent awoke for the final time; (7) time the respondent left their bed; (8) perceived sleep quality; and (9) an additional space for open-ended comments [23]. There is also an extended Consensus Sleep Diary and a pediatric Consensus Sleep Diary. Sleep diaries are especially useful because of their capacity to detect multiple sleep factors, highlighting their utility in studies interested in multidimensional sleep health. For example, from a Consensus Sleep Diary completed over multiple weeks, one can extrapolate patterns of stability or irregularity in sleep duration, bed timing, wake timing, and mid-sleep timing.

Sleep Quality QuestionnairesPittsburgh Sleep Quality Index

The Pittsburgh Sleep Quality Index (PSQI), one of the most frequently used subjective measures of sleep quality [19], was designed to measure patients’ global sleep quality and sleep disturbances over the past month. The PSQI is not specific to any one sleep disorder; rather, it captures sleep quality regardless of sleep disorder status. The PSQI is a reliable method of differentiating between “good” and “bad” sleepers [25]. There are 19 self-reported questions and five optional bed partner questions that cover seven components [26]. These components include subjective sleep quality, SOL, TST, habitual SE, sleep disturbances, use of sleep medication, and daytime dysfunction [25]. The total score ranges from zero to 21, with higher scores reflecting poorer sleep quality [26]. A global score higher than five indicates a “poor” sleeper [27]. Sub-scores of the PSQI can be calculated for individual components of sleep quality (e.g., TST, SOL, SE, daytime dysfunction) [28]. Like sleep diaries, the PSQI also captures sleep timing (i.e., bed timing, wake timing, and sleep onset latency), though it does not offer insights into the regularity of that sleep timing over the 1-month period being queried.

RU-SATED

The Satisfaction, Alertness, Timing, Efficiency, and Duration (SATED) Scale is designed to measure these components within a multi-dimensional sleep health framework. SATED approaches sleep health with a positive reference frame and emphasizes the positive impact of good sleep on overall health [4]. The SATED has been modified to include sleep regularity and is now called the RU-SATED. The RU-SATED includes the following statements measured on a Likert scale: (1) I go to sleep and wake up at about the same time every day, (2) I sleep 7–9 h per night, (3) The middle of my sleep period is between 2:00 am and 4:00 am, (4) I am awake for less than 30 min between the time I go to bed and the time I get out of bed, and (5) I stay awake all day without dozing, and (6) I am satisfied with my sleep. This measure is psychometrically valid [29].

Sleep Hygiene Index

Sleep hygiene refers to the practices and behaviors that produce the best environment for quality sleep to occur. Sleep hygiene is strongly related to sleep quality and modestly related to daytime sleepiness [30, 31]. Some common sleep hygiene tips include maintaining a regular sleep and wake schedule (even on the weekends), avoiding stimulants like caffeine and nicotine, and establishing a sleep routine. The Sleep Hygiene Index is a 13-item self-report questionnaire that quantifies a person’s current sleep hygiene practices. The Sleep Hygiene Index has a good internal reliability (Cronbach’s α = 0.66) and a high test-retest reliability (r(139) = 0.71, p < 0.01). It is also highly correlated with the PSQI (described above) and the Epworth Sleepiness Scale (ESS; described below) [30].

Circadian Chronotype Questionnaires

Chronotype is a circadian-driven preference for certain bedtime and waketimes and is typically categorized as morningness, intermediate, or eveningness. Chronotype reflects someone’s natural preference for when the person feels most awake, sleepiest, and what times a person prefers to sleep. There are specific questionnaires designed to assess chronotype [22].

Morningness-Eveningness Questionnaire

The Morningness-Eveningness Questionnaire (MEQ) is used to identify one’s preferred time of day to complete various daily activities and consists of 19 questions, using both Likert and time scales [22]. The scores range from 16 to 86: 16–30 is definitely evening type, 31–41 is moderately evening type, 42–58 is intermediate type, 59–69 is moderately morning type, and 70–86 is definitely morning type. This questionnaire was developed by researchers James A. Horne and Olov Östberg in the 1970s and was first validated using participants between 18 and 32 years old [32]. The full-scale internal consistency is 0.82 [32]. The MEQ is highly correlated (r = −0.70, p < 0.001) with dim light melatonin onset (DLMO), the gold standard objective method for determining circadian phase [33]. The MEQ has also been validated for children and adolescents [34, 35], and has been translated for several languages [36,37,38].

Munich Chronotype Questionnaire

The Munich Chronotype Questionnaire (MCTQ) assesses chronotype using the sleep midpoint on free or non-workdays. Sleep timing on non-workdays is highly influenced by the individual’s circadian clock, and could help identify chronotype [22]. The MCTQ contains 17 items and four distinct categories, including work schedule, workday sleep schedule, free day sleep schedule, and self-assessment of chronotype [39]. The questions gather information about sleep times, fully awake times, SOL, and sleep inertia (i.e., temporarily reduced cognitive and sensory-motor function that can occur shortly after waking) [40]. A study comparing the validity of the MCTQ with wrist actigraphy found a strong correlation between the two measurements with an r value of 0.73 and a p-value of < 0.001 [41]. The MCTQ has also been modified for shift workers (MCTQShift) due to the additional difficulties shift workers often experience with sleep [22, 42]. There is also an ultra-short version of the MCTQ, the µMCTQ, which reduces it from 17 items to 6, and has been externally validated with DLMO [43].

It is worth emphasizing the distinction in chronotype assessment between the MEQ and MCTQ. The MEQ utilizes personal preference whereas the MCTQ utilizes sleep behavior (i.e., sleep midpoint) on free or non-workdays. Theoretically, sleep on free or non-workdays is sleep closer to an individual’s preferred time. However, due to schedule constraints and life circumstance (e.g., childcare of young children), this may not always be the case. Researchers will need to consider their population of interest when choosing between these measures. However, we caution researchers from using a sleep diary or portions of the PSQI on sleep timing to infer a person’s chronotype. Researchers are instead recommended to include a validated measure specific to chronotype such as the MEQ or MCTQ. If a researcher is interested in proxies of circadian misalignment, then the use of a sleep diary in conjunction with a circadian questionnaire like the MEQ may provide some insight into how aligned or misaligned a person’s sleep is compared to what their preferred (i.e., circadian preference) sleep would be.

Clinical Sleep Disorder Questionnaires

Insomnia is one of the most studied sleep disorders due to its impact. An estimated one third of U.S. adults report insomnia symptoms, with 4–26% meeting clinical diagnostic criteria [44]. Insomnia diagnostic criteria include trouble falling asleep, trouble staying asleep, or waking up too early for ≥ 3 nights/week for ≥ 3 months duration [19]. Another common sleep disorder is OSA. OSA has an estimated prevalence between 3 and 17%, which is even greater among men and older age groups [45]. OSA prevalence has increased over time with the upward trend in obesity and poses an increased risk for developing CVD [46]. This disorder is characterized by changes in sleep-related breathing that reduce or block airflow in the upper airways during sleep [47]. Early diagnosis and treatment may help reduce risk of comorbidities that are associated with unmanaged OSA, including hypertension [48]. Additionally, RLS—also known as Willis-Ekbom Disease—affects a significant share of the population, with estimates ranging from 3 to 11% [49]. Given the prevalence of these disorders and their impact on CVD-related outcomes, it is critical to consider the inclusion of screeners for insomnia, OSA, and RLS in sleep and cardiovascular research.

Insomnia Severity Index

The Insomnia Severity Index (ISI) is the most widely used clinical questionnaire for insomnia screening. It assesses sleep onset, sleep maintenance, early morning awakenings, satisfaction with sleep patterns, interference with daily functioning, and degree of concern about sleep problems [50]. The ISI items coincide with both the DSM (version IV) and International Classification of Sleep Disorders (10th edition) diagnostic criteria for insomnia [51]. The ISI internal consistency had a Cronbach’s α score of 0.92, a sensitivity of 82.4%, and specificity of 83.1% for detecting clinical insomnia in a primary care sample (n = 410) [51]. The ISI is also sensitive to change, making it an ideal measure for both behavioral and pharmacologic clinical trials of insomnia treatment [52]. ISI scores range from 0 to 28, with 7, 14, 21, and 28 marking the upper bounds of no clinically significant insomnia, subthreshold (mild) insomnia, moderate clinical insomnia, and severe clinical insomnia. In other words, a score of 15 or higher is often used as a threshold for clinically meaningful insomnia [50].

PROMIS Sleep Disturbance and Sleep-Related Impairment Short Forms

The National Institutes of Health developed the Patient-Reported Outcomes Measurement Information System (PROMIS) program using item-response theory with the goal of creating forms that more accurately measure numerous self-report constructs of interest than previously available questionnaires [53]. There are two sleep-related PROMIS measures, the Sleep Disturbance form and the Sleep-Related Impairment form. These forms are 8-item questionnaires with Likert responses ranging from one to five. There are adult and pediatric versions of both forms. Raw scores are converted into T-Scores with higher scores indicating greater sleep disturbance.

STOP-Bang Questionnaire

The STOP-Bang questionnaire was designed specifically to meet the need for a quick and easy-to-score screening tool for OSA [54]. The name of this questionnaire is an acronym that relates to common OSA features: snoring, tiredness, observed apnea, elevated BP, elevated body mass index, increased age, greater neck circumference, and male sex [54]. STOP-Bang scores range from 0 to 8 with 3–4 indicating an intermediate risk of OSA, while 5 and higher indicates high risk of OSA. The STOP-Bang questionnaire was created for adults over the age of 18 and has been validated against polysomnography (PSG), the objective gold standard for diagnosing OSA [55]. In a meta-analysis that examined data from over 17 studies and included nearly 10,000 people, the sensitivity for this questionnaire to correctly identify someone with risk of moderate-to-severe and severe OSA was at 94% and 96%, respectively [55].

Berlin Questionnaire

The Berlin Questionnaire was also created to screen for OSA [56]. The Berlin Questionnaire has 11 items that cover three sub-categories: snoring, daytime sleepiness, and history of hypertension [57]. If two or more sub-categories are endorsed, there is high risk for OSA. The Berlin Questionnaire has a high internal consistency (Cronbach correlations of 0.86 to 0.92) and has been validated against portable sleep monitoring machines such as PSG [56].

Cambridge-Hopkins Diagnostic Questionnaire

The Cambridge-Hopkins Diagnostic Questionnaire is an 11-item self-report measure designed to categorize people into “RLS,” “non-RLS,” and “possible RLS.” It consists of two core features of RLS and nine other, similar symptoms that are used in the differential diagnosis of RLS. It demonstrated 87.2% sensitivity and 94.4% specificity for diagnosing RLS [58].

International RLS Rating Scale

The International RLS Rating Scale is a 10-item self-report measure designed to assess the severity of RLS symptoms over the past week among those with an established RLS diagnosis. Its Likert responses range from 0 to 4. Examples of questions are: “Overall, how would you rate the RLS discomfort in your legs or arms?” and “How severe was your sleep disturbance due to your RLS symptoms?” [59] Scores range from 0 to 40, with 0–10 indicating mild RLS, 11–20 indicating moderating RLS, 21–30 indicating severe RLS, and 31–40 indicating very severe RLS.

Epworth Sleepiness Scale

Excessive daytime sleepiness is defined as difficulty maintaining alertness and wakefulness during the wake phase of the 24-hour sleep-wake cycle [19]. Some symptoms that correlate with excessive daytime sleepiness include trouble staying awake, trouble focusing, difficulty making decisions, and memory problems [60]. Excessive daytime sleepiness is one of the most prevalent symptoms of someone with an underlying sleep disorder such as OSA, narcolepsy, or idiopathic hypersomnia [61].

The ESS is used to identify excessive daytime sleepiness in both clinical and research settings [52], and it offers a standardized and cost-effective way to measure sleepiness in people who may have a sleep disorder [19]. This questionnaire asks about the likelihood of falling asleep during normal daily activities (e.g., watching television, driving, reading) to distinguish between sleepiness and fatigue [52]. Each daily activity has been chosen based on differing levels of “somnoficity,” a term referring to posture, activity, and the likelihood of that situation resulting in dozing [52]. Scores range from 0 to 24, with 0–5 indicating lower normal sleepiness, 6–10 indicating higher normal sleepiness, 11–12 indicating mild excessive daytime sleepiness, 13–15 indicating moderate excessive daytime sleepiness, and 16–26 indicating severe excessive daytime sleepiness.

Karolinska Sleepiness Scale

The Karolinska Sleepiness Scale was developed as a one-dimensional one-question sleepiness scale that is sensitive to change, as it asks about the 10 minutes prior to completing the questionnaire [62]. Respondents rate their current sleepiness on a 9-point Likert scale. A score of 7 (“sleepy, but no effort to stay awake”) or higher may indicate impaired performance. It has been widely used in studies about shift work, jet lag, driving abilities, attention, and performance [63]. Since it measures state rather than trait sleepiness, it is not used in clinical settings [62]. A study involving 16 female participants between 33 and 43 years of age required them to complete the Karolinska Sleepiness Scale eight times daily for three consecutive days [63]. There was a moderate correlation (r = 0.56) between higher Karolinska Sleepiness Scale and increased reaction times, number of lapses, alpha and theta power density, and alpha attenuation coefficients [63]. The Karolinska Sleepiness Scale results were also compared with other behavioral variables and electroencephalogram (EEG), and had high validity for measuring sleepiness [63]. The Karolinska Sleepiness Scale has also been validated in a Spanish-speaking Colombian population [64].

Stanford Sleepiness Scale

The Stanford Sleepiness Scale, like the Karolinska Sleepiness Scale, evaluates one’s state sleepiness at the time of measure administration with a 7-point Likert scale [65]. The Stanford Sleepiness Scale also includes just one item, asking the respondent to rate their sleepiness at that specific time [66]. Thus, it is best used when repeated throughout a research study or over the course of treatment to track changes [65]. Scores of 5–7 are generally considered functionally impaired. To evaluate the Stanford Sleepiness Scale, five college students were given the questionnaire for five consecutive days and were sleep deprived on the fifth day [65]. After sleep deprivation, scores for the Stanford Sleepiness Scale increased significantly, which correlated with predicted levels of sleepiness [66]. The Stanford Sleepiness Scale was moderately correlated to sleep deprivation among college students (r = 0.474) [66].

Functional Outcomes of Sleep

The Functional Outcomes of Sleep Questionnaire (FOSQ) is used to determine the effects of fatigue on daytime activities and quality of life among individuals with a sleep disorder [67]. It has both long (30 items) and short (10 items) versions [