PSQI (Pittsburgh Sleep Quality Index): Take It, Score It, Understand Your Results
The PSQI (Pittsburgh Sleep Quality Index) is a 19-item self-report questionnaire that assesses your sleep quality and disturbances over the past month. Developed by Buysse and colleagues at the University of Pittsburgh in 1989, it remains one of the most widely used sleep quality measures in clinical practice and sleep research worldwide. The PSQI produces seven component scores plus a global 0-21 score, and a global score above 5 is widely used as a threshold for poor sleep quality that warrants clinical follow-up. The PSQI is a screening tool, not a diagnosis of any specific sleep disorder.
What is the PSQI?
The PSQI (Pittsburgh Sleep Quality Index) is a sleep quality questionnaire developed by Dr. Daniel Buysse and colleagues at the University of Pittsburgh and first published in 1989. It was designed to be a brief, valid, and reliable measure of sleep quality and disturbance suitable for both clinical and research use. Unlike a single-question rating of 'how well did you sleep,' the PSQI breaks sleep quality into seven measurable components and combines them into a global score. It has been validated in many populations and translated into dozens of languages, and it remains the most cited self-report sleep quality measure in the world.
What the PSQI measures
The PSQI measures sleep over the past month across seven components: (1) subjective sleep quality — how well-rested you feel; (2) sleep latency — how long it takes you to fall asleep; (3) sleep duration — how many hours you actually sleep; (4) habitual sleep efficiency — time asleep as a percentage of time in bed; (5) sleep disturbances — how often things like waking up, nightmares, pain, or breathing problems interrupt sleep; (6) use of sleeping medication — prescription or over-the-counter; and (7) daytime dysfunction — trouble staying awake or staying motivated during the day. Each component is scored 0-3 and the seven combine into a global score from 0 to 21.
How the PSQI is administered
The PSQI is a self-report questionnaire that takes about 5 to 10 minutes to complete. It contains 19 items that you answer yourself plus 5 additional items that a bed partner or roommate may answer (these are used for clinical information but do not affect your score). The 19 self-rated items cover your usual bed time, time to fall asleep, wake time, hours slept, and the frequency of various sleep problems. The recall window is the past month, which is intentionally longer than for some other screeners because sleep quality is meaningful over weeks rather than days. No special preparation is required; the validity of the score depends on honest, considered answers.
Who uses the PSQI
The PSQI is used routinely in sleep medicine clinics, primary care, psychiatry, geriatrics, and clinical research. It serves both as an initial screen for sleep problems that may warrant further evaluation — such as a sleep study, an insomnia assessment, or screening for sleep apnea — and as an outcome measure to track sleep quality over time during treatment. Outside clinical care, the PSQI is one of the most-used sleep measures in research on chronic disease, mental health, occupational health, athletics, and aging. Its broad use means scores can often be compared across studies and patient groups.
PSQI is a screening tool, not a diagnosis
The PSQI provides an overall measure of sleep quality, but it does not by itself diagnose any specific sleep disorder such as insomnia disorder, obstructive sleep apnea, restless legs syndrome, or a circadian rhythm disorder. A global PSQI score above 5 is widely used as a threshold for poor sleep quality and is a signal to talk with a clinician, not a clinical verdict. A score at or below 5 does not rule out a sleep disorder either — some disorders, particularly sleep apnea, can be present without the person noticing poor sleep quality. If sleep problems are affecting your day-to-day life, a clinician can determine whether a sleep study or other evaluation is warranted.
How to score and interpret your results
The PSQI turns your answers into seven component scores, each from 0 to 3, then adds those seven numbers into a global PSQI score from 0 to 21. A 0 on any component means no trouble in that area; a 3 means severe trouble. Higher global totals point to poorer sleep quality. A global score greater than 5 is the cutoff most often cited as a sign of poor sleep.
The cutoff comes from the original 1989 validation, where a global score above 5 correctly classed up to 89% of patients as poor sleepers. It is a screening threshold, not a diagnostic line. A score of 6 does not mean you have a sleep disorder, and a score of 4 does not rule one out — the PSQI is a screening tool, not a diagnosis.
Global PSQI score bands
The bands below restate the published evidence in plain English. The only sharply defined cutoff in the validation data is the >5 threshold.
| Global score | What it suggests | What to do |
|---|---|---|
| 0–5 | Generally in line with acceptable sleep quality on this screener | No specific action needed from this result alone |
| 6–10 | Above the >5 cutoff — suggests poor sleep quality and a need for follow-up | Discuss with a clinician, especially if you notice daytime impact |
| 11–21 | Well above the cutoff — suggests serious ongoing sleep trouble | Bring the score to a clinician; a sleep study may be warranted |
If your global score is 5 or 6, treat it as borderline. The PSQI’s validation showed high but imperfect sensitivity and specificity at this threshold (about 90% / 87%), so a small share of people on either side of the line will be misclassed. Trends over time and daytime symptoms matter as much as the single number. The PSQI was also designed as an outcome measure, so repeating it after a clinician-guided change in care can tell you more than a single snapshot.
The 7 PSQI components in detail
Each of the seven PSQI components captures a different side of sleep quality. A high score on one points to a different talk than a high score on another. The table below sums up what each component measures and what an elevated score might suggest — not what it diagnoses. The PSQI is a screening tool, not a diagnosis.
What each component captures
| Component | What it measures | What a high score might suggest |
|---|---|---|
| 1. Subjective sleep quality | Your overall self-rating of how well you slept over the past month | A general sense that sleep is unrefreshing; useful to compare against the more behavioral components |
| 2. Sleep latency | How long it takes you to fall asleep after going to bed | Trouble starting sleep, a hallmark complaint in insomnia symptoms |
| 3. Sleep duration | The number of hours you actually sleep per night | Habitually short sleep relative to adult targets of 7–9 hours |
| 4. Habitual sleep efficiency | Time asleep as a percentage of time in bed | A mismatch between time in bed and time asleep — often seen with broken or maintenance insomnia symptoms |
| 5. Sleep disturbances | How often things like waking up, breathing problems, pain, or nightmares break up sleep | Several sources of nighttime disturbance worth noting for a clinician |
| 6. Use of sleeping medication | How often you use prescription or over-the-counter sleep aids in the past month | Reliance on sleep aids worth a clinician review — this component tracks less tightly with the rest in some studies |
| 7. Daytime dysfunction | Trouble staying awake during the day and trouble staying motivated | Daytime fallout of poor sleep, including weaker thinking, slower reaction time, and lower mood |
The sleep-medication and daytime-dysfunction components track less tightly with the other five in some later reviews. That is why researchers debate whether the PSQI is best modeled as one factor or as two or three. In practice, clinicians look at single component scores next to the global score. A person whose high score is driven by sleep latency and disturbances has a different clinical picture than someone whose score is driven by daytime dysfunction alone. The 5 partner-rated items covering snoring, breathing pauses, leg twitching, and confusion are not part of the scored components but are collected for clinical context.
What a high PSQI score means — and what it doesn’t
A global PSQI score above 5 means that your self-rated sleep quality, on this screener, is poorer than the validation sample’s good sleepers. It suggests that sleep is worth raising with a clinician, and depending on other symptoms, it may be a reason to consider a sleep study. The PSQI is a screening tool, not a diagnosis.
A high PSQI score does not identify which sleep condition is driving it. The PSQI does not diagnose insomnia disorder, obstructive sleep apnea, restless legs syndrome, or any circadian rhythm disorder by itself. Those diagnoses require a clinician’s review, and several call for added testing such as a sleep study.
What the PSQI can miss
A score at or below 5 does not rule out a sleep disorder. Some conditions — most notably sleep apnea — can be present without the person being aware of poor sleep quality, since brief nighttime arousals may not register as remembered awakenings. A person with witnessed breathing pauses and loud snoring can have a near-normal PSQI and still benefit from a sleep study referral. Restless legs syndrome is only partly captured — the partner-rated leg-twitching item gives some signal, but it is not part of the scored components. Circadian rhythm disorders, where total sleep may be enough but timing is off, may not push the PSQI clearly above the cutoff either.
A related symptom-focused tool is the Insomnia Severity Index, which zooms in on insomnia symptoms and rounds out the broader PSQI snapshot.
How accurate is the PSQI?
The PSQI was validated in 1989 by Buysse and colleagues at the University of Pittsburgh using three groups: 52 healthy sleepers, 54 depressed patients, and 62 sleep-disorder patients. In that study, a global score above 5 produced 89.6% sensitivity and 86.5% specificity for telling good from poor sleepers, with a kappa of 0.75. The PMC review cites rounded figures of about 90% sensitivity and 87% specificity, and notes the cutoff of 5 correctly flagged up to 89% of patients.
Reliability evidence is also strong. The original work showed acceptable internal consistency, test-retest reliability, and overall validity. Later studies report Cronbach’s alpha of 0.83 for internal consistency and test-retest reliability of r = 0.85. The PSQI also shows agreement with other measures, with notable correlations to sleep diaries and depression rating scales.
Real-world reach
The PSQI’s reach in the literature is unusual for a self-report tool. The University of Pittsburgh notes the 1989 paper has been cited in over 34,000 peer-reviewed articles; the PMC review puts the count at about 37,666 on Google Scholar. The tool has been translated into 56 added languages, and the PMC review puts the total at over 60.
Population studies using PSQI > 5 show steady estimates across very different cohorts. A German community study of 9,248 adults found a 36% rate of poor sleep, with women reporting more problems than men. A Korean nationwide sample of 165,193 adults showed a 41% rate, tied to lower income, poor health habits, and worse mental health. The same cutoff holds up well across these very different settings.
Limitations and considerations
The PSQI is well-validated but has built-in limits. The most important is the most obvious: the PSQI is a subjective measure. It captures how you perceive your sleep over the past month, and that view can differ from objective sleep recording. This is what lets mild sleep apnea slip past a near-normal PSQI, when the person does not consciously notice the arousals.
A second point concerns the questionnaire’s structure. Later reviews have found that a one-factor model often fits poorly, and that two- or three-factor models tend to fit better. Sleep medication use and daytime dysfunction track less tightly with the other components in some studies. The global score remains a useful measure of overall sleep quality at the group level, while single components may better serve detailed clinical review.
Design constraints to know
A third limit is the one-month recall window. The month-long frame steadies the measure against any single bad week, but it also limits how well it picks up acute changes. A recent stressor disrupting only the past five nights may not move the global score much. The recall window cannot be shortened or stretched without breaking the tool — the PSQI is only validated for the one-month frame.
Finally, the PSQI was developed and validated in adult populations and should not be stretched to children. Modifying any part of the tool requires written permission from the University of Pittsburgh, which owns the PSQI. Validated translations in 56+ languages have received Linguistic Validation Certificates confirming conceptual equivalence, so a validated translation is preferable to an ad-hoc one.
What to do with your results: when to talk to a clinician
The PSQI helps you decide whether to bring sleep into a talk with a clinician. The PSQI is a screening tool, not a diagnosis, and the score is most useful when read alongside how your sleep is affecting your day. Lasting sleep problems can be a sign of conditions like insomnia or sleep apnea, both of which warrant a professional review and may involve a sleep study.
When the result warrants follow-up
Consider talking to a clinician — and bringing your PSQI score — if any of these apply:
- Your global PSQI score is above 5, especially if it stays above 5 on a repeat assessment a few weeks later
- You feel lasting daytime drowsiness, trouble focusing, or slowed reaction time that affects driving, working, or studying
- A bed partner has noted loud snoring with breathing pauses at night, even if your PSQI score is near or below the cutoff
- Sleep trouble has lasted more than a few months and is affecting your mood, work, or relationships
- You are using sleep medications (prescription or over-the-counter) on a regular basis and want clinician input — captured by component 6
- A high score is driven by sleep latency or sleep disturbances and you also notice signs of anxiety or depression — these often benefit from joint review
Poor or short sleep raises the risk of heart and breathing problems and affects metabolism and thinking, so lasting poor sleep is worth taking seriously even when no acute event prompts the visit. US sleep medicine practice is guided by the American Academy of Sleep Medicine.
Sleep hygiene as a starting concept
Outside a clinical visit, MedlinePlus lists several broadly recommended sleep-habit steps. These include steady bed and wake times, avoiding afternoon and evening caffeine, getting regular exercise but not near bedtime, keeping the bedroom cool, dark, and quiet, and using the bed only for sleep. These are general ideas, not personal prescriptions. If simple habit changes do not bring relief within a few weeks, that itself is a reason to escalate to a clinician. Sleep medication choices are best discussed with a clinician.
Frequently asked questions
How many items are on the PSQI?
The PSQI has 19 self-rated items plus 5 added items answered by a bed partner or roommate. Only the 19 self-rated items count toward your score. The 5 partner items collect things like snoring or breathing pauses for clinical context and are not scored.
What is a good PSQI score?
On the PSQI’s 0–21 global scale, lower scores mean better sleep quality, and a global score of 5 or below is generally in line with acceptable sleep on this screener. A score greater than 5 suggests poor sleep quality and warrants follow-up. The PSQI is a screening tool, not a diagnosis.
How is the PSQI scored?
Each of the seven components is scored 0 to 3, where 3 means the most trouble. The seven component scores are then added into a global score from 0 to 21. The components themselves are listed in the next question.
What are the 7 components of the PSQI?
The seven components are subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medication, and daytime dysfunction. Each is rated 0–3, and together they form the 0–21 global score. The structure has been used steadily since the 1989 validation.
How do I cite the PSQI (Buysse 1989)?
The standard citation is: Buysse, D.J., Reynolds, C.F., Monk, T.H., Berman, S.R., & Kupfer, D.J. (1989). The Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research. Psychiatry Research, 28(2), 193–213. The PMID is 2748771 and the DOI is 10.1016/0165-1781(89)90047-4.
Can I use the PSQI for free?
The PSQI may be reprinted without charge for non-commercial research and educational purposes, and academic or US-government-funded studies use it free of charge. Commercial, clinical, and certain publication uses are subject to licensing fees set by the University of Pittsburgh, which owns the tool.
How long does the PSQI take to complete?
The PSQI is brief enough to complete in a single sitting. The validated recall window is the past month, so honest, careful answers about typical sleep over that window matter more than how quickly you finish.
Is the PSQI available in other languages?
Yes. The PSQI has been translated into 56 added languages through steps that produce a Linguistic Validation Certificate confirming conceptual equivalence to the original. A peer-reviewed review puts the total at over 60 languages. Using a validated translation is preferable to a self-made one.