Generalised Anxiety Disorder-7: Scoring, Cutoffs & Interpretation

01 - What it measures

What this scale measures.

The GAD-7 measures the severity of generalised anxiety symptoms over the past two weeks. Each of the seven items maps onto a core feature of generalised anxiety - feeling nervous or on edge, not being able to stop or control worrying, worrying too much about different things, trouble relaxing, restlessness, irritability, and a sense that something awful might happen. Respondents rate how often each symptom has bothered them on a 0 to 3 scale, and the items are summed to a total of 0 to 21.

The scale captures a single underlying anxiety dimension rather than discrete diagnoses, so it is best read as a measure of overall symptom load. It is sensitive to change, which makes it useful for tracking response to treatment within a patient over time. It is not designed to differentiate generalised anxiety from panic, social anxiety, or obsessive-compulsive presentations.

02 - Origin & purpose

Where it comes from.

Spitzer, Kroenke, Williams and Löwe published the GAD-7 in Archives of Internal Medicine in 2006. The existing anxiety instruments were either too long for routine use, such as the Hamilton Anxiety Rating Scale and the Beck Anxiety Inventory, or too narrow, such as the Penn State Worry Questionnaire. The team derived seven items from a 13-item pool against the DSM-IV criteria for generalised anxiety disorder and validated them across 2,740 patients in primary care.

They established cutoffs of 5, 10, and 15 for mild, moderate, and severe symptoms, with the threshold of 10 or above carrying 89% sensitivity and 82% specificity against a structured interview. Löwe and colleagues standardised the measure in the general population in 2008, confirming a single-factor structure and population norms. Toussaint and colleagues established its sensitivity to change and a minimal clinically important difference of around 4 points in 2020. The scale is in the public domain and free to use for clinical, research, and commercial purposes when the 2006 paper is cited.

03 - Scoring & cutoffs

How scoring works.

Each of the seven items is rated 0 to 3 for the last two weeks, and the total is the simple sum of all items, with a range of 0 to 21. None of the items is reverse-scored. The bands are 0 to 4 minimal, 5 to 9 mild, 10 to 14 moderate, and 15 to 21 severe.

The threshold of 10 or above is the most-cited cutoff for likely generalised anxiety disorder; some primary-care guidelines use 8 or above to widen the screening net. An eighth item asks how difficult these problems have made daily functioning - it is recorded but not added to the total.

04 - Validation evidence

How well it performs.

In the original validation against a blinded MHP-administered structured interview, the cutoff of ≥10 produced 89% sensitivity and 82% specificity for generalised anxiety disorder. Subsequent independent samples have replicated those properties closely, with sensitivity in the 0.83–0.92 range and specificity 0.78–0.84 across primary care, psychiatric outpatient, and obstetric populations.

Internal consistency is high (Cronbach's α ≈ 0.92) and one-week test–retest reliability sits at 0.83, supporting use as a repeated measure. Factor analyses consistently recover a single underlying anxiety factor, and the scale is sensitive to change with intervention.

05 - How it compares

How it compares to the alternatives.

06 - When to use it

Right tool, wrong tool.

The GAD-7 is intended as a brief screen and severity measure, not a diagnostic test. In clinical settings it suits pre-visit triage in primary care or therapy intake, setting a referral threshold in stepped-care models, and tracking response to treatment every 2 to 4 weeks. For self-screening, it can help a person decide whether their worry warrants a conversation with a clinician.

A positive screen should always be followed by a clinical interview against DSM-5 or ICD-11 criteria. The GAD-7 performs adequately in ages 13 and above; for children under 13, use the SCARED screener instead.

07 - Confidence & precision

Reading the score with care.

A single GAD-7 score should be read as an interval rather than an exact point. The standard error of measurement is about 1.4 points, so small fluctuations between visits sit within measurement noise. A change of 4 points or more is generally treated as clinically meaningful. Ceiling effects can appear at scores of 18 and above, which limits sensitivity to change at the top of the range.

08 - Limitations

What it cannot tell you.

The GAD-7 is a screening and severity tool, not a diagnostic instrument, and cannot replace a clinical interview. It does not differentiate generalised anxiety from panic, social anxiety, or obsessive-compulsive disorder, so a positive screen needs follow-up to identify the specific condition.

It is not validated for children under 13. Somatic items can be confounded by physical illness, and self-report is open to under- or over-reporting. Ceiling effects at the severe end reduce its usefulness for tracking already-extreme presentations.

Common questions from clinicians

FAQ

References

Source literature.

1. [1]Spitzer RL, Kroenke K, Williams JBW, Löwe B A brief measure for assessing generalized anxiety disorder: the GAD-7 (2006)
2. [2]Löwe B, Decker O, Müller S, et al. Validation and standardization of the GAD-7 in the general population (2008)
3. [3]Toussaint A, Hüsing P, Gumz A, et al. Sensitivity to change and minimal clinically important difference of the GAD-7 (2020)
4. [4]Plummer F, Manea L, Trepel D, McMillan D Screening for anxiety disorders with the GAD-7: a diagnostic meta-analysis (2016)