The Bergen Insomnia Scale (BIS) is a brief, 6-item self-report screening measure of insomnia symptoms over the past month for adults aged 18+. The items ask how many days a week each night-time and daytime insomnia symptom occurs, based on diagnostic symptom criteria.
The BIS has been used in hospital medical settings with treatment-seeking patients with sleep apnoea (Lundetræ et al., 2021) and early postpartum women (Takenoshita et al., 2026). It has also been used in outpatient mental health settings with adults with attention-deficit/hyperactivity disorder (Brevik et al., 2017) or recurrent depression (Evanger et al., 2024). Each item counts the days a week a symptom occurs, so the BIS shows whether insomnia symptoms are present and how severe they are.
The BIS supports several clinical uses:
The night-time group is trouble falling asleep, staying asleep, and waking too early, and the daytime group is not feeling rested, daytime impairment, and dissatisfaction with sleep (Pallesen et al., 2008). The groupings broadly reflect the DSM-IV-TR emphasis on problems sleeping at night and their effects during the day (Pallesen et al., 2008). The two groups are moderately related, so the BIS gives one total score (Pallesen et al., 2008). The groupings can be useful to describe patterns or clusters of a client’s symptoms. The report’s response table follows the DSM-5-TR: night-time symptoms (items 1 to 3), daytime impairment and dissatisfaction with sleep (items 5 and 6), and not feeling rested (item 4) on its own, because the DSM-5 no longer uses it as a criterion (Bjorvatn et al., 2018).
The BIS produces a total score and a count of symptom categories. Higher scores mean more insomnia symptoms, or symptoms on more days.
The descriptor combines the DSM-5 algorithm with a total-score threshold of 14. The algorithm is at least one night-time symptom (items 1 to 3: trouble falling asleep, staying asleep, or waking too early) with daytime impairment or dissatisfaction with sleep (item 5 or 6), each on 3 or more days a week. The BIS does not assess the DSM-5-TR 3-month duration.
There are four possible descriptions in a report:
At the first administration, the report shows the total score (0 to 42) against two samples from the development study (Pallesen et al., 2008): community adults and patients referred to a sleep clinic. Simplified curves show where each sample’s scores mostly fall. Dashed lines mark each sample’s 50th percentile, the median (7.9 and 21.1), estimated from its published average and SD (Pallesen et al., 2008). A bar along the x-axis shades from white to red as scores rise, as a visual guide to severity. A grey dashed line marks the total-score threshold of 14. From the second administration, the report plots the total score over time (0 to 42) with the same line, then each item’s days a week (0 to 7) with a line at the 3-day item threshold.
A change of 10 or more points in the total score is a significant reduction or increase in insomnia severity, and a change of fewer than 10 points may be interpreted with caution. The threshold is a reliable change index calculated from the community sample’s SD (9.73) and internal consistency (alpha = .87; Pallesen et al., 2008). The report compares the total with the first administration and, from the third, with the most recent one.
Pallesen et al. (2008) developed and validated the BIS using the DSM-IV-TR criteria for insomnia. Items were scaled as days a week over the past month, and night-time items as nights with more than 30 minutes awake, to align with clinical diagnostic criteria (Lacks & Morin, 1992). The past-month form, which NovoPsych uses, was validated in Norway in students (n = 320), community adults aged 18 to 80 (n = 2,645), and sleep clinic patients (n = 225; Pallesen et al., 2008).
The BIS shows good internal consistency (alpha .79 in students, .87 in community adults, and .80 in sleep clinic patients; Pallesen et al., 2008). Its stability over 2 weeks was adequate (test-retest r = .77 in 182 students; Pallesen et al., 2008). These results demonstrate the reliability of the full scale in indexing insomnia symptoms.
In students on the past-month form, the total correlated strongly with the Athens Insomnia Scale (r = .79) and the Pittsburgh Sleep Quality Index (r = .73; Pallesen et al., 2008). It correlated less with depression (Beck Depression Inventory-II, r = .55) and anxiety (Beck Anxiety Inventory, r = .32). Against a diagnostic interview in 144 women 6 weeks after childbirth, the total had moderate accuracy for insomnia disorder (AUC = .73; form version not stated; Takenoshita et al., 2026). These results support convergent and discriminant validity, and moderate accuracy for insomnia disorder.
The BIS captures change across administrations during treatment. In a randomised trial of internet-delivered cognitive behavioural therapy for insomnia, the total fell from an estimated 25.85 to 14.94 over the 9-week program (n = 95; Hagatun et al., 2019), a large within-group effect (d = −1.50; form version not stated). In 442 sleep apnoea patients starting continuous positive airway pressure, the total fell from 18.78 to 12.77 after an average of 23.1 weeks, a medium effect on a mix of past-month and 3-month forms (d = 0.65; Lundetræ et al., 2021). Changes in the BIS total score can track change in insomnia severity.
The BIS developers count a symptom as present when it occurs on 3 or more days a week. Based on the DSM-IV-TR, they indicate insomnia when at least one night-time sleep symptom and one daytime symptom reach that level (Pallesen et al., 2008). The symptoms and the 3-day threshold remain consistent with the DSM-5-TR criteria. The DSM-5-TR no longer includes not feeling rested, and it extended the required duration from 1 to 3 months (American Psychiatric Association, 2022; Bjorvatn et al., 2018).
More community adults meet the BIS algorithm (24.9% to 33.7%; Bjorvatn et al., 2023; Vaag et al., 2016) than European population studies report for insomnia disorder (5.7% to 19%; Porcheret et al., 2024). Self-report scales count symptoms but do not assess every criterion, such as adequate opportunity to sleep or another condition with the same symptoms (Porcheret et al., 2024). Using only the algorithm may overestimate insomnia disorder.
The total score assesses severity. The developers printed community percentiles only from the 80th percentile up, the top fifth of community adults, close to the highest reported rate of insomnia disorder (Pallesen et al., 2008; Porcheret et al., 2024). Women with untreated insomnia disorder in an interview study had a median total of 32, above the 95th percentile (Takenoshita et al., 2026). A high total therefore shows symptoms at the frequency seen in insomnia disorder.
The algorithm shows whether the symptoms fit the diagnostic criteria, and the total shows how often they occur, so the two can disagree. Reading both gives a clearer indication of insomnia disorder than either alone, so NovoPsych has developed a descriptor that combines both:
No. “Consistent with insomnia” means the DSM-5 algorithm is met and the total score is 14 or more, so the symptoms fit the diagnostic criteria and occur often. A diagnosis also depends on how long the problem has lasted, 3 months or more in the DSM-5-TR, and whether there is enough opportunity to sleep, which the BIS does not ask about (American Psychiatric Association, 2022; Pallesen et al., 2008).
Every question asks about the past month. The authors chose a month to match the time criterion for insomnia in the DSM-IV-TR and to keep the time frame short (Pallesen et al., 2008). Some people answer for the past week instead (Pallesen et al., 2008), so reminding the client that the questions cover the whole month keeps the answers on the right time frame.
The BIS measures insomnia, which differs from the sleepiness that comes with sleep apnoea or other disorders of excessive sleep. In sleep clinic patients, higher BIS scores went with less daytime sleepiness, and patients with sleep apnoea reported fewer days of trouble falling asleep than patients with other sleep problems (Pallesen et al., 2008). When a client is sleepy by day rather than unable to sleep, the BIS may not capture their main sleep problem.
Each item asks about the past month, so administrations about a month apart describe separate periods. The items show which symptoms are improving, and a symptom that drops from 3 or more days a week to fewer no longer counts as present. In a trial of mindfulness-based cognitive therapy for recurrent depression, participants reported fewer days with trouble falling asleep and night waking than a waitlist group after treatment, but no difference in early waking, not feeling rested, daytime impairment, or dissatisfaction (Evanger et al., 2024). During treatment for sleep apnoea, the daytime items fell more than the night-time items (Lundetræ et al., 2021). Symptoms that stay at 3 or more days a week can point to what still needs attention in treatment.
The authors published percentiles only from the 80th up, separately for men and women in four age bands (Pallesen et al., 2008). They cannot be combined into one set for all adults, and nothing below the 80th can be read from them. The report reads each item against the authors’ threshold of 3 or more days a week, and the total against a threshold of 14, the lowest 80th percentile among the eight sex and age groups (Pallesen et al., 2008). The first-administration chart shows where the total sits against simplified and estimated curves for community adults and sleep clinic patients, with medians estimated from the published averages and SDs.
The BIS was developed in adults, and NovoPsych offers it for adults aged 18 and over. Studies in adolescents used changed forms: a Turkish translation reworded the first item to mention phones and tablets (Bay & Ergun, 2018), and a Norwegian study asked about the past 3 months rather than the past month (Arnesen et al., 2025). Those findings do not apply to the adult past-month form.
Pallesen, S., Bjorvatn, B., Nordhus, I. H., Sivertsen, B., Hjørnevik, M., & Morin, C. M. (2008). A new scale for measuring insomnia: The Bergen Insomnia Scale. Perceptual and Motor Skills, 107(3), 691–706. https://doi.org/10.2466/pms.107.3.691-706
American Psychiatric Association. (2022). Diagnostic and statistical manual of mental disorders (5th ed., text rev.). https://doi.org/10.1176/appi.books.9780890425787
Arnesen, I. B., Bjorvatn, B., Pallesen, S., Waage, S., Gradisar, M., Wilhelmsen-Langeland, A., & Saxvig, I. W. (2025). Insomnia in adolescent epidemiological studies: To what extent can the symptoms be explained by circadian factors? Chronobiology International, 42(1), 58–69. https://doi.org/10.1080/07420528.2024.2444677
Bay, T., & Ergun, A. (2018). Validity and reliability of Bergen Insomnia Scale (BIS) among adolescents. Clinical and Experimental Health Sciences, 8, 268–275.
Bjorvatn, B., Berge, T., Lehmann, S., Pallesen, S., & Saxvig, I. W. (2018). No effect of a self-help book for insomnia in patients with obstructive sleep apnea and comorbid chronic insomnia – A randomized controlled trial. Frontiers in Psychology, 9, Article 2413. https://doi.org/10.3389/fpsyg.2018.02413
Bjorvatn, B., Waage, S., Pallesen, S., Buysse, D. J., & Saxvig, I. W. (2023). The association between different sleep health dimensions and sex, age, education, circadian preference, and chronic insomnia: A representative population-based study. Sleep Advances, 4(1), zpad041. https://doi.org/10.1093/sleepadvances/zpad041
Brevik, E. J., Lundervold, A. J., Halmøy, A., Posserud, M. B., Instanes, J. T., Bjorvatn, B., & Haavik, J. (2017). Prevalence and clinical correlates of insomnia in adults with attention-deficit hyperactivity disorder. Acta Psychiatrica Scandinavica, 136(2), 220–227. https://doi.org/10.1111/acps.12756
Evanger, L. N., Flo-Groeneboom, E., Sørensen, L., & Schanche, E. (2024). Mindfulness-based cognitive therapy improves insomnia symptoms in individuals with recurrent depression: Secondary analyses from a randomized controlled trial. Frontiers in Psychiatry, 14, Article 1231040. https://doi.org/10.3389/fpsyt.2023.1231040
Hagatun, S., Vedaa, Ø., Nordgreen, T., Smith, O. R. F., Pallesen, S., Havik, O. E., Bjorvatn, B., Thorndike, F. P., Ritterband, L. M., & Sivertsen, B. (2019). The short-term efficacy of an unguided internet-based cognitive-behavioral therapy for insomnia: A randomized controlled trial with a six-month nonrandomized follow-up. Behavioral Sleep Medicine, 17(2), 137–155. https://doi.org/10.1080/15402002.2017.1301941
Lacks, P., & Morin, C. M. (1992). Recent advances in the assessment and treatment of insomnia. Journal of Consulting and Clinical Psychology, 60(4), 586–594. https://doi.org/10.1037/0022-006X.60.4.586
Lundetræ, R. S., Saxvig, I. W., Aurlien, H., Lehmann, S., & Bjorvatn, B. (2021). Effect of continuous positive airway pressure on symptoms and prevalence of insomnia in patients with obstructive sleep apnea: A longitudinal study. Frontiers in Psychology, 12, Article 691495. https://doi.org/10.3389/fpsyg.2021.691495
Pallesen, S., Bjorvatn, B., Nordhus, I. H., Sivertsen, B., Hjørnevik, M., & Morin, C. M. (2008). A new scale for measuring insomnia: The Bergen Insomnia Scale. Perceptual and Motor Skills, 107(3), 691–706. https://doi.org/10.2466/pms.107.3.691-706
Porcheret, K., Hopstock, L. A., & Nilsen, K. B. (2024). Prevalence of insomnia in a general adult population cohort using different diagnostic criteria: The seventh survey of the Tromsø study 2015–2016. Sleep Medicine, 119, 289–295. https://doi.org/10.1016/j.sleep.2024.05.002
Takenoshita, M., Jalali-Sohi, A., Hidajat, N. Z., Michel, G., Guo, N., Manber, R., Lyell, D. J., Barwick, F., Morris, A. M., Moody, H., Elkhateb, R., Mhyre, J., Cella, D., Shaunfield, S., Mackey, S., Tian, L., Carmichael, S. L., Panelli, D. M., Tang, X., & Sultan, P. (2026). Insomnia disorder at 6 weeks postpartum: A prospective study. Journal of Clinical Sleep Medicine, 22(1), Article 152. https://doi.org/10.1007/s44470-026-00150-3
Vaag, J., Saksvik-Lehouillier, I., Bjørngaard, J. H., & Bjerkeset, O. (2016). Sleep difficulties and insomnia symptoms in Norwegian musicians compared to the general population and workforce. Behavioral Sleep Medicine, 14(3), 325–342. https://doi.org/10.1080/15402002.2015.1007991
Senior Research Fellow and Clinical Psychologist