ISI (Insomnia Severity Index): Take It, Score It, Understand Your Results
The ISI (Insomnia Severity Index) is a brief 7-item self-report questionnaire that measures how severe your insomnia symptoms have been over the past two weeks. Developed by Morin and colleagues at Université Laval, it is one of the most widely used insomnia outcome measures in sleep medicine, clinical research, and primary care worldwide. The ISI takes about 5 minutes to complete and produces a 0-28 score that helps a clinician decide whether further evaluation is warranted. The ISI is a screening tool, not a diagnosis of any specific sleep disorder: results should be reviewed with a healthcare professional.
What is the ISI?
The ISI (Insomnia Severity Index) is a brief insomnia questionnaire developed by Dr. Charles Morin and colleagues at Université Laval in Quebec, originally published in the early 1990s and re-validated against current diagnostic criteria by Morin and colleagues in 2011. It is designed to be a simple, valid, and reliable measure of perceived insomnia severity, suitable for both clinical screening and outcome tracking during treatment. The ISI has been translated into more than 20 languages and is one of the most cited insomnia self-report measures in the world, used routinely in sleep medicine clinics, primary care, and clinical trials.
What the ISI measures
The ISI measures perceived insomnia severity across seven items that map closely onto current diagnostic criteria for insomnia disorder. The items cover problems falling asleep, problems staying asleep, problems waking too early, satisfaction with current sleep pattern, how noticeable your sleep problems are to others, how worried you feel about your sleep problems, and how much your sleep difficulties interfere with daily functioning (such as tiredness, mood, ability to perform at work or school, concentration, and quality of life). By covering both the nighttime symptoms and the daytime consequences, the ISI captures what clinicians actually want to know when assessing insomnia.
How the ISI is administered
The ISI is a self-report questionnaire that takes about 5 minutes to complete. Each of the 7 items is rated on a 0-4 scale, with 0 meaning no problem and 4 meaning very severe problem (the exact anchor words vary slightly by item). The recall window is the past two weeks. You can complete it on paper, on a screen, or with a clinician reading the items aloud. The total score is the simple sum of the seven items, with a possible range from 0 to 28. No special preparation is required; honest, considered answers about your sleep over the past two weeks are what give the score its meaning.
Who uses the ISI
The ISI is used routinely in sleep medicine clinics, primary care, psychiatry, and clinical research as both an initial screening instrument and an outcome measure to track how someone's insomnia is changing during treatment. It is one of the recommended instruments in research and clinical trials evaluating cognitive behavioral therapy for insomnia (CBT-I), pharmacological interventions, and other sleep-related treatments. Outside formal care, the ISI is also used in public health surveys and occupational research. Its brevity and broad use mean ISI scores can often be compared across studies and patient groups.
ISI is a screening tool, not a diagnosis
The ISI gives a brief snapshot of perceived insomnia severity, but it does not by itself diagnose insomnia disorder or any other specific sleep condition such as obstructive sleep apnea, restless legs syndrome, or a circadian rhythm disorder. A high ISI score is a signal that further conversation with a clinician is warranted — not a clinical verdict. A low score does not rule out a sleep disorder, particularly if you have signs that a bed partner has noticed (such as breathing pauses or loud snoring) but you have not. If sleep problems are affecting your day-to-day life, a clinician can decide whether further evaluation, such as a sleep study, makes sense.
How to score and interpret your results
The ISI total score is the simple sum of the seven items, each rated on a 0-4 scale, giving a possible range of 0 to 28. Higher numbers reflect more severe self-reported insomnia symptoms over the recall window. Most clinicians and the original validation work use four interpretive bands to translate that raw number into a severity category.
The bands and the clinically used cutoff scores come directly from the Morin 2011 validation study (community sample N=959; clinical sample N=245).
| ISI total | Severity band | Notes |
|---|---|---|
| 0-7 | No clinically significant insomnia | Symptoms in the normal range |
| 8-14 | Subthreshold insomnia | Some symptoms, below clinical threshold |
| 15-21 | Moderate clinical insomnia | At or above the traditional >=15 cutoff |
| 22-28 | Severe clinical insomnia | Highest band; warrants prompt clinician review |
| Cutoff >=15 | Sensitivity 47.7-78.1%, specificity 98.3-100% | High specificity, lower sensitivity |
Alternate cutoffs by setting
The traditional >=15 cutoff is a high-specificity threshold designed to confidently identify moderate-to-severe cases. For broader screening, two lower cutoffs perform better. In a community sample, a score of 10 was optimal, with 86.1% sensitivity and 87.7% specificity for detecting insomnia cases. In a clinical sample of people already presenting to a sleep service, a cutoff of 11 yielded 97.2% sensitivity and 100% specificity for diagnosed insomnia.
Which threshold matters depends on the setting. A primary care or self-screening context usually benefits from the lower cutoff (around 10), because the goal is to catch most people with meaningful sleep problems even at the cost of some false positives. A specialist confirming severity for treatment planning may rely on the 15+ band, where a positive result is very unlikely to be a false alarm. The ISI is a screening instrument across all of these uses, not a diagnostic test.
The 7 ISI items in detail
The seven items map closely onto current diagnostic criteria for insomnia disorder and cover both nighttime symptoms and daytime impact. Below is a plain-English description of what each item is asking about. The verbatim wording is copyrighted by the instrument’s developer and must be obtained through a licensed source — what follows describes the construct, not the official phrasing.
- Item 1 — Sleep onset difficulty. How hard it has been to fall asleep at the start of the night.
- Item 2 — Sleep maintenance. Difficulty staying asleep, including waking during the night.
- Item 3 — Early morning awakening. Waking earlier than intended and being unable to return to sleep.
- Item 4 — Sleep satisfaction. How satisfied or dissatisfied you currently feel with your sleep pattern (a reverse-scored satisfaction item).
- Item 5 — Daytime interference. How much sleep problems are getting in the way of daily functioning — tiredness, mood, concentration, work or school performance, quality of life.
- Item 6 — Noticeability to others. How apparent your sleep difficulties seem to people around you.
- Item 7 — Distress about sleep. How worried or distressed you feel about your sleep problems.
Why the daytime items matter
Three of the seven items (interference, noticeability, distress) deliberately shift the focus away from the bedroom and toward the consequences of poor sleep. Current diagnostic frameworks treat daytime impairment as a core feature of insomnia, not an optional add-on. A person who sleeps poorly but reports no daytime distress and no functional impact will tend to score lower than someone with similar nighttime symptoms plus significant daytime cost. That design choice is one reason the ISI tracks treatment response well — successful treatment usually improves both halves of the construct, not just the hours-of-sleep half.
What a high ISI score means — and what it doesn’t
A score at or above 15 is a meaningful signal that your sleep difficulties are in the moderate-to-severe range as you experience them. It is a reason to have a structured conversation with a clinician about evaluation and possible treatment. It is not, by itself, a diagnosis of insomnia disorder, and it is not a diagnosis of any other specific sleep condition. The Morin 2011 validation describes the ISI as an instrument “to detect cases of insomnia” and to monitor treatment response — detection is not diagnosis.
The ISI is a subjective self-report measure. It asks how you perceive your sleep, not what is happening physiologically while you sleep. Several common sleep-related conditions are difficult or impossible to detect with a brief perceived-severity questionnaire:
- Breathing-related sleep problems. Snoring, witnessed pauses in breathing, and oxygen drops during sleep are typically noticed by a bed partner or measured during a formal sleep study; the ISI does not ask about them.
- Leg-movement and physical-sensation symptoms. Uncomfortable leg sensations or repetitive limb movements during sleep are not covered by ISI items.
- Sleep-timing problems. A mismatch between when you can sleep and when you need to be awake (for example with night-shift work or long-distance travel) can feel like insomnia but reflects timing rather than sleep difficulty per se.
- Behaviors during sleep. Sleepwalking, talking, acting out dreams, or unusual nighttime behaviors are not captured by ISI items.
A low ISI score also does not rule out a sleep disorder. If a bed partner has noticed loud snoring, gasping, breathing pauses, or kicking and jerking, those observations matter regardless of what your ISI total suggests, and they are reasons to ask a clinician about a formal sleep evaluation. The ISI tells you something useful about your sleep experience — it does not tell you everything about your sleep. The ISI remains a screening instrument throughout, not a diagnostic test.
How accurate is the ISI?
The ISI’s psychometric performance has been examined across community and clinical samples and is generally considered strong for a brief self-report measure.
Reliability
Internal consistency — the degree to which the seven items measure a coherent construct — was excellent in both validation samples: Cronbach’s alpha was 0.90 in the community sample and 0.91 in the clinical sample. Item-total correlations ranged from 0.55 to 0.81, with the satisfaction and worry items showing the strongest correlations to the total score.
Validity
Convergent validity was demonstrated through correlations with established measures of sleep quality, mood, anxiety, and fatigue.
| Comparison measure | Correlation with ISI total |
|---|---|
| Pittsburgh Sleep Quality Index (PSQI) | r = 0.80 |
| Beck Depression Inventory (BDI) | r = 0.48-0.50 |
| Beck Anxiety Inventory (BAI) | r = 0.48 |
| Multidimensional Fatigue Inventory | r = 0.20-0.55 |
The very strong correlation with the PSQI (r = 0.80) is expected — the two instruments measure overlapping constructs from slightly different angles. Criterion validity was also supported: 89.7% of people reporting sleep problems scored at or above 15, and 98.3% of those reporting no sleep problem scored below 8.
Sensitivity to treatment change
The ISI is sensitive enough to track improvement during treatment. In the Morin 2011 analysis, a reduction of more than 7 points corresponded to a moderate improvement rating by independent assessors, and a reduction of more than 8 points corresponded to a marked improvement. Average change scores were -8.4 points for moderate improvement and -9.9 points for marked improvement. That responsiveness is one reason the ISI is widely used as an outcome measure in clinical trials.
Limitations and considerations
The ISI is well-validated, but it is a brief subjective instrument and has real bounds you should keep in mind when interpreting your score. It remains a screening tool, not a diagnostic test.
- Recall window. The original Morin 2011 validation specifies a recall period of the past month. Some clinical adaptations and digital deployments use a shorter two-week recall to align with other mood and sleep measures. Both versions are in use; the band cutoffs were established against the longer recall period.
- Subjective measure. The ISI captures perception of sleep, not measured sleep. People with similar physiology can score quite differently depending on distress, attention to sleep, and daytime impact.
- Limited coverage of other sleep problems. The instrument focuses on perceived insomnia severity and does not ask about snoring, breathing pauses, leg discomfort, sleep timing, or behaviors during sleep, so other sleep-related conditions can be present in someone with a low score.
- Item-level discrimination varies. Item response theory analysis in the clinical sample showed five of seven items had adequate-to-excellent discriminative capacity, but item 1 (sleep onset) had weaker discrimination in that middle-aged clinical group. This is a research-level nuance, not a reason to drop the item, but it does mean a low score on item 1 alone should not be over-interpreted.
- Cross-cultural adaptations. The ISI has been used in many translated forms, and performance can vary across cultural and demographic contexts; cutoffs validated in one population should not be assumed to transfer perfectly to another.
- Comorbid mood and anxiety symptoms. Because depression and anxiety correlate moderately with ISI scores (BDI r = 0.48-0.50, BAI r = 0.48), an elevated ISI in someone with significant mood symptoms may partly reflect those overlapping experiences.
The ISI also differs in structure from companion sleep instruments such as the PSQI. The PSQI is a broader sleep-quality questionnaire that asks about multiple aspects of recent sleep behavior; the ISI focuses on perceived severity and impact. Clinicians sometimes use both because they capture complementary information.
What to do with your results — when to talk to a clinician
The ISI is a screening tool, not a diagnosis, and your score is most useful as a starting point for a conversation with a healthcare professional. Use these score-based prompts as a guide for that conversation, not as an instruction set.
Talk to a clinician if any of the following apply to your situation:
- ISI total of 15 or higher. This corresponds to the moderate-to-severe band and meets the traditional clinical cutoff (specificity 98.3-100%). A clinician evaluation is appropriate.
- ISI total of 10-14 with meaningful daytime impact. Lower scores can still warrant clinical attention, particularly when item 5 (daytime interference) and item 7 (distress) are elevated. The community-screening cutoff is 10 (sensitivity 86.1%, specificity 87.7%).
- Sleep difficulty for 3 or more nights per week, lasting 3 months or more. This duration pattern aligns with the framework for chronic insomnia and is worth raising with a clinician regardless of your exact score.
- A bed partner has noticed loud snoring, gasping, witnessed pauses in breathing, or unusual movements during sleep. These observations may suggest a sleep-related condition that the ISI does not assess and that may need a sleep study.
- Sleep problems coincide with low mood, anxiety, or significant life stress. Stress, depression, and emotional distress are recognized risk factors for insomnia and are treatable.
- Sleep difficulties are causing safety risks. Drowsy driving and accidents are documented serious risks of sleep deprivation.
What evaluation and treatment can involve
A clinician will typically ask about sleep habits, take a medical and medication history, and may request a sleep diary. A sleep study may be recommended when indicated. Recognized treatment categories for insomnia include lifestyle and sleep-schedule changes, cognitive behavioral therapy, and medication. The American Academy of Sleep Medicine publishes separate clinical practice guidelines for behavioral and psychological treatment of chronic insomnia and for pharmacologic treatment of chronic insomnia. Specific medication selection and dosing decisions are clinician-led and outside the scope of any screening result.
Frequently asked questions
Is the ISI the same as the PSQI?
No. The ISI is a 7-item insomnia severity measure focused on perceived severity and impact. The Pittsburgh Sleep Quality Index is a longer sleep-quality questionnaire that covers multiple components of recent sleep behavior. The two correlate strongly (r = 0.80) but capture different information.
How is the ISI scored?
You sum the seven items, each rated 0-4, giving a total from 0 to 28. The original interpretive bands are: 0-7 no clinically significant insomnia, 8-14 subthreshold insomnia, 15-21 moderate insomnia, and 22-28 severe insomnia. The ISI remains a screening measure, not a diagnostic test.
What is a normal ISI score?
Scores in the 0-7 band reflect an absence of clinically significant insomnia symptoms in the original validation framework. There is no single “normal” number — people sleep and perceive sleep differently — but scores in that band are typically not flagged as a clinical concern. A clinician can put your score in context.
Can I use the ISI to diagnose myself with insomnia?
No. The ISI is a screening instrument that helps quantify perceived insomnia severity; it is not a diagnostic test. Diagnosis of insomnia disorder is made by a clinician using a fuller history and current diagnostic criteria, often supplemented by a sleep diary or other tools.
How often should I repeat the ISI?
There is no fixed schedule. The ISI is sensitive enough to track treatment response, with reductions of more than 7 points indicating moderate improvement and more than 8 points indicating marked improvement on independent assessor ratings. People in active treatment commonly repeat it periodically; outside of treatment, repeating it whenever your sleep changes meaningfully is a reasonable approach.
Does the ISI cover sleep apnea?
No. The ISI does not include items on snoring, breathing pauses, or oxygen-related symptoms. A clinician can refer you for a sleep study when a breathing-related sleep condition is suspected.
What is the difference between subthreshold and moderate insomnia on the ISI?
A subthreshold score (8-14) reflects some insomnia symptoms that fall below the traditional clinical cutoff and may still warrant attention if daytime impact is significant. A moderate score (15-21) meets the traditional cutoff and is a stronger signal that clinical evaluation is appropriate.
Is the ISI free to use?
The instrument is copyrighted, and licensing is managed through the developer’s distribution channel; clinicians and researchers typically obtain it through licensed sources rather than copying it informally. Re-publishing the verbatim items without permission is not permitted. The ISI remains a screening tool — your score should be reviewed with a clinician.