Written and maintained by CASRAI Editorial Board
Last updated
General information about a research instrument, not medical advice. A questionnaire score does not diagnose a sleep disorder; sleep concerns belong with a clinician.
Short answer
The Pittsburgh Sleep Quality Index (PSQI) is a self-rated questionnaire that assesses sleep quality and disturbances over a 1-month interval. Its 19 items generate seven component scores that sum to one global score, and Buysse and colleagues’ 1989 validation found that a global score greater than 5 separated good from poor sleepers with 89.6% sensitivity and 86.5% specificity. That cutoff came from one sample mix and is not universal, as the table below shows.
Definition and scope
The PSQI was introduced in Buysse DJ, Reynolds CF 3rd, Monk TH, Berman SR, Kupfer DJ, The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research, Psychiatry Research 1989;28(2):193-213 (doi 10.1016/0165-1781(89)90047-4). The authors built it because few questionnaires had been designed to measure sleep quality in clinical populations, particularly among psychiatric patients. NIDA Data Share describes it as a brief, validated instrument that measures sleep quality, with 19 items, a one-month time frame and the same seven components.
The seven components are:
- subjective sleep quality
- sleep latency
- sleep duration
- habitual sleep efficiency
- sleep disturbances
- use of sleeping medication
- daytime dysfunction
How the PSQI is used in practice
Respondents answer about their usual sleep over the past month. The component scores are summed into a global score; the Shirley Ryan AbilityLab Rehabilitation Measures summary (a secondary source) gives the global range as 0 to 21, with higher scores indicating worse sleep quality, and estimates 5 to 10 minutes to administer and about 5 minutes to score. The 1989 validation studied 52 healthy “good” sleepers, 54 depressed patients and 62 sleep-disorder patients over 18 months, and reported acceptable internal homogeneity, test-retest reliability and validity. Researchers who need the general vocabulary can read psychometrics and test-retest reliability.
Cutoff scores differ by population
Most pages repeat “greater than 5 means poor sleep”. The published validation work summarised by the Rehabilitation Measures page shows the threshold and its accuracy depending on the sample and language version. Sensitivity and specificity trade off as the cutoff moves, a pattern explained in sensitivity vs. specificity.
| Sample (study) | Cutoff | Sensitivity / specificity |
|---|---|---|
| Good sleepers, depressed and sleep-disorder patients (Buysse et al. 1989) | > 5 | 89.6% / 86.5% |
| Primary insomnia (Backhaus et al. 2002) | > 5 | 98.7% / 84.4% |
| Primary insomnia (Backhaus et al. 2002) | > 6 | 93.4% / 100% |
| Chinese version, primary insomnia vs. controls (Tsai et al. 2005) | > 5 | 98% / 55% |
| Chinese version, primary insomnia vs. controls (Tsai et al. 2005) | > 6 | 90% / 67% |
| Japanese version, primary insomnia (Doi et al. 2000) | 5.5 | 85.7% / 86.6% |
| Japanese version, major depression (Doi et al. 2000) | 5.5 | 80% / 86.6% |
| Patients with cancer (summary page) | > 8 | 0.90 / 0.83 |
Source for all rows except the first: Shirley Ryan AbilityLab Rehabilitation Measures, PSQI page, read 2026-11-07; the first row is the PubMed abstract. The practical point for a protocol is that a cutoff should be cited together with the population and language version it was validated in. When a questionnaire is administered in another language, see translation and back-translation of research instruments.
Permissions and licensing checklist for researchers
The Rehabilitation Measures page reproduces the University of Pittsburgh’s statement. Based on that text:
- The copyright is owned by the University of Pittsburgh.
- The form may be reprinted without charge only for non-commercial research and educational purposes.
- No changes or modifications to the form may be made without prior written permission from the University of Pittsburgh.
- Permission requests go through the University of Pittsburgh (Center for Sleep and Circadian Science) or Daniel J. Buysse, MD.
- Commercial use, or commercially sponsored research, requires licensing through the University of Pittsburgh Office of Technology Management.
This is why searches for an editable “PSQI Word document” or a modified or translated copy raise a permission question rather than only a formatting one. An edited, shortened or re-ordered version is a modification under the quoted statement.
Where it matters for labs and clinics
In study design the PSQI is one of the patient-reported measures an investigator chooses for sleep outcomes; the selection steps are covered in choosing and validating a patient-reported outcome measure. For comparison with another widely used self-report screener, see the PHQ-9 and GAD-7 guides. NIDA Data Share lists the PSQI as a data-sharing instrument, so studies using it can map their variables to a shared definition.
Common misunderstandings
Is a global score above 5 a diagnosis?
No. The 1989 paper reports the score’s accuracy in distinguishing good from poor sleepers in a validation sample; it describes utility in psychiatric clinical practice and research, not a stand-alone diagnosis.
Does it measure last night’s sleep?
No. It asks about a 1-month interval, so it is not a measure of a single night or a daily diary.
Is 5 the cutoff for every group?
No. See the table: the studies above used 5, 5.5, 6 or 8 depending on population and language version.
Can the questionnaire be freely edited?
Not under the University of Pittsburgh statement quoted above, which requires prior written permission for changes.
FAQ
What does PSQI stand for?
Pittsburgh Sleep Quality Index.
How many items and components does it have?
Nineteen individual items generate seven component scores, which sum to one global score (Buysse et al. 1989).
What is the score range?
0 to 21, with higher scores indicating worse sleep quality, per the Rehabilitation Measures summary.
Is it free to use?
A free reprint is stated for non-commercial research and educational purposes only; commercial use requires licensing, as set out in the checklist above.
Sources
- Buysse DJ et al. Psychiatry Research 1989;28(2):193-213, PMID 2748771, PubMed abstract, read 2026-11-07.
- NIDA Data Share: Pittsburgh Sleep Quality Index, read 2026-11-07.
- Shirley Ryan AbilityLab Rehabilitation Measures: Pittsburgh Sleep Quality Index (secondary summary of the cited studies and the University of Pittsburgh copyright statement), read 2026-11-07.
