The General Health Questionnaire screens for current psychological distress and possible common mental disorder - the non-psychotic conditions, chiefly depression and anxiety, seen most often in primary care. Its items ask whether the respondent has recently experienced breaks in normal functioning: loss of concentration, sleep disturbance, strain, low mood and reduced confidence, each judged against how things usually are for that person over the past few weeks.
It is deliberately a state measure rather than a trait measure: it detects recent change from a person's own baseline rather than long-standing characteristics. The family includes 60-, 30-, 28- and 12-item versions. The GHQ-12 is by far the most widely used internationally; the GHQ-28 adds four scaled subscales (somatic symptoms, anxiety and insomnia, social dysfunction, and severe depression) for research settings that want a symptom profile.
02 - Origin & purpose
Where it comes from.
The GHQ was developed by Sir David Goldberg and introduced in his 1972 monograph The Detection of Psychiatric Illness by Questionnaire as a 60-item instrument for detecting probable psychiatric 'caseness' in general medical settings. The scaled 28-item version followed in 1979 (Goldberg & Hillier), and the 12-item short form was consolidated in Goldberg & Williams' 1988 User's Guide.
It became one of the most widely used mental-health screening instruments in the world, translated into dozens of languages and validated in the WHO's 15-centre study of mental illness in general health care. It remains a staple of occupational, epidemiological and primary-care research, though its licensing requirements have pushed many clinical services towards free alternatives such as the Kessler scales and the PHQ family.
03 - Scoring & cutoffs
How scoring works.
Each item has four response options. Two scoring methods are in use. The original GHQ method scores responses 0-0-1-1, so each item is effectively binary; for the GHQ-12 this gives a range of 0-12, with scores of 3 or more (some settings use 2 or more, others 4 or more) indicating probable psychological distress warranting assessment. The Likert method scores responses 0-1-2-3, giving a 0-36 range for the GHQ-12 that is better suited to measuring severity along a continuum; commonly cited thresholds sit around 12 or more, but they vary by population and purpose. Because the items ask about change from the person's usual state, someone with chronic, unchanging difficulties can score deceptively low with the GHQ method - a recognised weakness of the 0-0-1-1 approach.
04 - Validation evidence
How well it performs.
The GHQ-12's validity evidence is extensive and international. In the WHO study of mental illness in general health care, conducted across 15 centres, the GHQ-12 discriminated cases of mental disorder from non-cases with sensitivity of roughly 84% and specificity of roughly 79%, performing as well as the longer GHQ-28. Reliability is consistently good: a 2024 reliability-generalisation meta-analysis reported pooled Cronbach's alpha estimates in the range 0.82-0.85 across studies.
~84%
SENSITIVITY
~79%
SPECIFICITY
α = 0.82-0.85
INTERNAL CONSISTENCY
12 items
LENGTH
05 - How it compares
How it compares to the alternatives.
Instrument
Items
Time
When to reach for it
K10
10
2-3 min
Free. Reach for it as the closest free substitute for general distress screening, with well-established severity bands.
Free with conditions. Reach for it to measure wellbeing rather than distress.
06 - When to use it
Right tool, wrong tool.
Reach for it when
-Research and occupational studies where GHQ continuity with the existing literature matters and a licence is budgeted.
-Screening for recent-onset psychological distress against a person's own baseline.
-Epidemiological work needing an extensively validated, internationally comparable distress measure.
-Settings that want the GHQ-28's four symptom subscales for a research profile.
Reach for something else when
-Any setting without a GL Assessment licence - the items may not be reproduced; use the K10, K6 or PHQ-4 instead.
-Diagnosing a specific disorder - the GHQ flags probable caseness, not which condition.
-Detecting chronic, unchanging difficulties - the change-from-usual wording can miss long-standing problems under GHQ scoring.
-Monitoring week-to-week treatment response against fixed severity bands - thresholds vary by population and scoring method.
-Assessing psychotic disorders or cognitive impairment.
07 - Confidence & precision
Reading the score with care.
A standard error of measurement is rarely reported for the GHQ-12 and no minimal clinically important difference has been established, so score changes between administrations should be interpreted cautiously. Cutoffs are also not universal: the optimal threshold shifts with the population's base rate of disorder and the scoring method used, which is why validation studies report thresholds ranging from 2 to 4 under GHQ scoring. Treat the score as an indicator of probable distress needing clinical assessment, not a measurement of severity change.
08 - Limitations
What it cannot tell you.
Licensed and fee-bearing: items cannot be reproduced without GL Assessment's written permission, which limits clinical adoption.
The change-from-usual response format can underscore chronic problems under the standard 0-0-1-1 method. Cutoffs vary across populations, languages and scoring methods, so no single threshold applies everywhere. The GHQ screens for probable caseness only - it cannot distinguish depression from anxiety or other conditions.
Its factor structure is debated (unidimensional versus multi-factor solutions driven by negatively worded items), which complicates subscale interpretation of the GHQ-12. It is not designed for, or validated in, psychotic or cognitive disorders.
09 - Licensing, explained
How licensing works.
Copyright David Goldberg (1978). Published and licensed by GL Assessment (part of GL Education), with distribution for research via Mapi Research Trust (ePROVIDE). The items may not be reproduced in any form - including photocopying - without the publisher's written permission, and per-use fees apply. This page is an information resource; the questionnaire itself is not shown.
Common questions from clinicians
FAQ
References
Source literature.
[1]Goldberg DP The Detection of Psychiatric Illness by Questionnaire. Oxford University Press, London (1972) ↩
See how Aisel removes friction where it costs most. A 20-minute walkthrough tailored to your clinic.
We value your privacy
We use cookies to analyse site usage and improve your experience. Analytics and embedded media (e.g. YouTube) only load if you accept. Read our cookie policy.