The Insomnia Severity Index (ISI): Validity, Scoring, and How to Access It

If you've been researching insomnia questionnaires, you've probably run into the Insomnia Severity Index (ISI) — the most widely used measure of its kind in sleep research and clinical practice. This page answers the specific questions people actually ask about it: who made it, how well-validated it is, what the score bands mean, whether you can get a copy, and whether a version exists for children.

What is the Insomnia Severity Index?

The ISI is a 7-item self-report questionnaire developed by Dr. Charles M. Morin, a psychologist and sleep researcher at Université Laval in Quebec, Canada. It was introduced in his 1993 book on the psychological assessment and management of insomnia, and formally validated as a research outcome measure a few years later.

Each of the 7 items is rated on a 0–4 scale, covering the severity of sleep-onset and sleep-maintenance difficulty, satisfaction with current sleep pattern, how noticeable the sleep problem is to others, how worried the person is about it, and how much it interferes with daily functioning. Scores are summed for a total ranging from 0 to 28.

Is the ISI valid and reliable?

Yes — it's one of the more extensively validated instruments in sleep medicine. The original validation study found solid internal consistency (a statistical measure of how well the 7 items hang together as one scale), with a reliability coefficient of about 0.74. A larger follow-up study by the same research group found reliability as high as 0.90 in some samples, and confirmed the measure reliably distinguishes people with insomnia from good sleepers and tracks treatment response over time — see the 2011 psychometric paper for the full analysis.

That 2011 paper is also the source of a detail worth knowing: statistically, a cutoff score of about 11 gave the best balance of sensitivity and specificity for detecting a clinical insomnia case in a general community sample (a slightly higher cutoff, around 14, performed best in samples already drawn from sleep clinics). This is a somewhat more precise, sample-dependent finding than the simpler four-band system most sites — including this one, until now — tend to quote.

ISI scoring and cutoffs explained

The scoring guide most commonly cited (including in Harvard Medical School's own hosted copy of the instrument) breaks the 0–28 total into four bands:

These four bands come from Harvard Medical School's own hosted copy of the ISI scoring guide.

Worth knowing: these bands are a convenient, widely-taught simplification, not the only statistically defensible way to interpret a score — see the note above about the 2011 study's sensitivity/specificity finding, which suggests the true clinical threshold sits a bit lower than the traditional 15 cutoff for the "moderate" band, at least in community (non-clinic) samples.

What does a score of 14, 15, or 16 mean?

This exact range gets searched a lot, probably because it's the awkward straddle point between two bands: 14 is the top of "subthreshold insomnia," while 15 and 16 fall into "moderate clinical insomnia" under the traditional four-band system. In practice, the difference between a 14 and a 15 is one Likert point on one question — it doesn't represent a meaningfully different clinical picture. If you're in this range, the more useful thing to look at is which specific items are driving your score (sleep-onset difficulty vs. sleep-maintenance vs. daytime interference), not which side of an arbitrary line the total falls on.

Is the ISI copyrighted? Can I download or print it?

Yes, it's copyrighted (© Charles M. Morin, 1993), and it isn't freely reproducible — including on websites like this one. It's distributed and licensed through Mapi Research Trust's ePROVIDE platform, which has separate request workflows depending on how it will be used (academic research without funding, funded academic research, and commercial use, among others).

That's actually why the two insomnia-related tools on this site don't reproduce the ISI's exact questions: we built our own original checkers instead, asking about similar territory in our own wording, rather than distributing a copyrighted clinical instrument without the appropriate license. If you specifically need the validated ISI itself — for research, clinical use, or just because you want the exact instrument researchers cite — request access directly through ePROVIDE, or ask your doctor or a sleep clinic, since many already hold a license.

Is there a version for children?

Not exactly the same instrument adapted downward — there's a genuinely separate, purpose-built measure. The Pediatric Insomnia Severity Index (PISI) is a 6-item parent-report questionnaire (with a self-report version for children old enough to complete it themselves), independently developed and validated by a different research team specifically for use in pediatric sleep evaluation.

A 2017 validation study found the PISI holds up well statistically and correlates as expected with other established measures of child sleep problems — a genuinely separate, purpose-built instrument, not a simplified adult ISI.

ISI scores and depression

Insomnia and depression are closely linked, and it shows up in ISI research: higher ISI scores correlate with more severe depressive symptoms, and the relationship runs in both directions — untreated chronic insomnia measurably raises the risk of developing major depressive disorder, and depression very commonly disrupts sleep in turn. Neither the ISI nor any self-report insomnia tool is designed to screen for depression specifically, so a high score on either front is worth mentioning to a doctor rather than interpreting in isolation.

A free alternative you can use right now

If you want a real ISI score — the one referenced in research papers and used by sleep clinics — that requires the actual licensed instrument. But if you just want a quick, honest read on your own sleep trouble right now, without waiting on a license request, we built two free tools that ask about similar territory in our own original questions:

The Insomnia Symptom Checker starts with frequency and duration (how many nights per week, how long it's been going on) before asking about pattern and daytime impact. The Insomnia Self-Assessment takes a different angle, focused on the habits and conditions most linked to insomnia risk — caffeine timing, screen use, schedule consistency, and sleep environment. Neither is a validated clinical instrument or a diagnosis; both are informal, genuinely useful starting points.

Frequently asked questions