The Hospital Anxiety and Depression Scale (HADS) is a 14-item self-report measure of anxiety and depression severity over the past week, built for medically ill patients. Seven items form an anxiety subscale (HADS-A) and seven a depression subscale (HADS-D), each scored 0-21. Its defining design choice is the deliberate exclusion of somatic symptoms - fatigue, insomnia, appetite change - so that scores are not inflated by the physical illness that brought the patient to hospital in the first place.
The depression subscale leans heavily on anhedonia, the loss of pleasure and interest, which its authors regarded as the core of depressive illness in the medically unwell. The HADS screens and tracks severity; it does not diagnose, and elevated scores call for clinical assessment.
02 - Origin & purpose
Where it comes from.
Anthony Zigmond and Philip Snaith published the HADS in 1983, having observed that existing depression questionnaires misclassified general-hospital outpatients: items about fatigue, weight and sleep scored the physical illness rather than the mood disorder. They wrote fourteen items free of somatic content, balanced the response scales, and validated the result in medical outpatient clinics in Leeds.
It became one of the most used measures in consultation-liaison psychiatry and health research worldwide, translated into well over a hundred languages. Copyright passed to GL Assessment (UK), with international distribution through the Mapi Research Trust; reproduction and electronic administration require permission and usually a per-use licence fee, which is why the questionnaire itself does not appear on this page.
03 - Scoring & cutoffs
How scoring works.
Each of the 14 items scores 0-3, summed separately into HADS-A and HADS-D subscale totals of 0-21; the two subscales are always reported separately and a combined total is discouraged. On either subscale, 0-7 is normal, 8-10 borderline, and 11-21 indicates probable clinical caseness. The screening literature generally uses 8 or above as the cutoff for possible disorder, and 11 or above as the stricter threshold for probable disorder. Scores reflect the past week, so the HADS is suitable for repeated administration during treatment or admission.
04 - Validation evidence
How well it performs.
The most-cited psychometric summary is the Bjelland et al. (2002) systematic review of 747 studies. At the common screening cutoff of 8 or above, HADS-A gave a sensitivity of 0.90 and specificity of 0.78; HADS-D at the same cutoff gave sensitivity 0.83 and specificity 0.79 - an unusually balanced screen at a single cutoff. The stricter cutoff of 11 raised specificity at the cost of sensitivity. Mean Cronbach's alpha across reviewed studies was 0.83 for HADS-A and 0.82 for HADS-D.
The factor structure has been debated throughout the scale's life: one, two and three-factor solutions have all been reported, which complicates subscale interpretation. A 2023 theoretical and methodological review (Lloyd et al.) reopened the structure question, though the two-subscale reading remains the clinical standard.
0.90
SENSITIVITY (HADS-A >=8)
0.78
SPECIFICITY (HADS-A >=8)
0.83
CRONBACH'S α, HADS-A
0.82
CRONBACH'S α, HADS-D
05 - How it compares
How it compares to the alternatives.
Instrument
Items
Time
When to reach for it
HADS
14
~5 min
14-item two-subscale screen (HADS-A and HADS-D, 0-21 each) built for medically ill patients; excludes somatic items. Licensed (GL Assessment / Mapi Research Trust).
10 items, free; can generate ICD-10/DSM diagnostic categories for depression, not just severity.
06 - When to use it
Right tool, wrong tool.
Reach for it when
-Screening for anxiety and depression in medically ill or hospital populations where somatic items would mislead.
-Oncology, cardiology and chronic-disease research with established HADS-based literatures.
-Repeated weekly severity tracking during treatment or admission, where the licence is already in place.
Reach for something else when
-Settings without a HADS licence - PHQ-9 and GAD-7 cover the same ground free.
-Suicide-risk assessment: the HADS has no suicidality item. Use PHQ-9 item 9 or a dedicated tool such as the ASQ or C-SSRS.
-Diagnosis: the HADS screens and tracks severity but cannot diagnose; a clinical interview is required.
-Severe depression in psychiatric settings, where the anhedonia-weighted HADS-D can under-detect.
07 - Confidence & precision
Reading the score with care.
Measurement precision differs by subscale: internal consistency is consistently good for HADS-A but more variable for HADS-D, so borderline HADS-D scores (8-10) deserve particular caution. No firmly established minimal clinically important difference exists for either subscale; a 1-2 point movement is within measurement error, and the borderline band exists precisely to absorb that uncertainty.
08 - Limitations
What it cannot tell you.
No somatic items means genuinely somatic presentations of depression can be missed, and the anhedonia-weighted HADS-D under-detects in some psychiatric samples. The factor structure has been debated for four decades (one, two and three-factor solutions all reported), which complicates subscale interpretation. There is no suicidality item. Licence fees limit use, especially electronic administration. Cutoffs were derived largely from Western hospital outpatient samples.
09 - Licensing, explained
How licensing works.
The HADS is copyrighted by GL Assessment and distributed internationally through the Mapi Research Trust (eprovide.mapi-trust.org); reproduction and electronic administration require permission and usually a per-use licence fee, so we cannot host the instrument here. PHQ-9 covers depression, GAD-7 covers anxiety, and PHQ-4 is the closest free equivalent to the HADS two-subscale design.
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.