Rating scales
Score the Generalised Anxiety Disorder-7: seven items, a 0–21 total, severity bands, and the ≥10 threshold at which further assessment is conventionally indicated.
Free · No sign-up · Runs entirely in your browser
Over the last two weeks, how often has the person been bothered by each problem? 0 = Not at all, 1 = Several days, 2 = More than half the days, 3 = Nearly every day.
Total score
0 / 21
Sum of all seven items
Severity band
Minimal
Spitzer et al., 2006
≥10 screen
Negative
Probable anxiety disorder
Enter the pre-treatment total to see change from baseline and the 50% response criterion.
Below the ≥10 threshold
Yes
Screening threshold, not remission
There is no remission figure here on purpose: unlike the HAM-D, MADRS and YMRS, the GAD-7 has no widely agreed remission cut-point, so the first reading above reports only whether the total has fallen below the ≥10 screening threshold — which is not the same claim. Response follows the ≥50% reduction criterion and 4 points is the change conventionally treated as clinically meaningful; both are reporting conventions, so state the definitions you applied.
Plain text: every item with its code and anchor, the total, the band, the threshold and — when a baseline is entered — the change from it.
Items, bands and the ≥10 cutoff follow Spitzer, Kroenke, Williams & Löwe (2006). This is a scoring aid for education and research support, not a diagnosis: the GAD-7 is a screening and severity measure, and a positive screen calls for a clinical assessment rather than standing in for one. Rate the items above, or load the example, to see the score update.
Reference · for the curious
The GAD-7 is seven items long, takes about a minute, costs nothing, and is the most widely used brief measure of anxiety in general practice. It was developed by Spitzer, Kroenke, Williams and Löwe in 2006 as a companion to the PHQ-9, using the same two-week window and the same 0–3 response format so that the two can be given together as one short questionnaire.
Each of the seven items is rated 0 (not at all), 1 (several days), 2 (more than half the days) or 3 (nearly every day) over the past two weeks, giving a total from 0 to 21. The conventional severity bands are 0–4 minimal, 5–9 mild, 10–14 moderate and 15–21 severe.
Ten or above is the usual threshold for further assessment. In the original validation study it gave a sensitivity of 89% and a specificity of 82% for generalised anxiety disorder — good enough to be useful as a screen, not good enough to diagnose.
What is less widely appreciated is that the GAD-7 is not specific to GAD. It performs reasonably as a screen for panic disorder, social anxiety disorder and post-traumatic stress disorder as well, which means a positive result indicates an anxiety disorder is likely somewhere in that family rather than that generalised anxiety specifically is present. Differentiating them requires a clinical assessment, not a higher cut-point.
Depression and anxiety co-occur often enough that screening for one alone misses a substantial share of cases. Because both instruments share a response format and a recall window, they combine into a single brief questionnaire at no extra cognitive cost to the respondent — and the ultra-brief PHQ-4 exists as exactly that pairing, two items from each.
If you are administering both, the PHQ-9 calculator scores the depression side, and its item 9 requires the separate attention described there.
The GAD-7 asks the respondent to summarise a fortnight in seven judgements. The HAM-A, by contrast, is a clinician-rated instrument with fourteen items and an explicit split between psychic and somatic anxiety, which gives it more resolution and makes it more sensitive to change — at the cost of a trained rater and fifteen to twenty minutes.
The two correlate strongly (around r ≈ 0.85 in reported comparisons) but they are not interchangeable, and converting between them is a rank-order approximation rather than a translation. Our comparison of GAD-7 versus HAM-A works through when the extra cost of a clinician-rated scale is worth paying, and clinician-rated versus self-report scales covers the general trade-off.
The GAD-7 is used repeatedly in practice and in trials. Response is commonly defined as a 50% reduction from baseline, and a change of around 4 points is often treated as clinically meaningful. As always, report the definition you used — response and remission rates are not comparable across studies that define the thresholds differently.
The GAD-7 is harder to rate from a transcript than the PHQ-9, and for a structural reason: six of its seven items are internal states — worrying, trouble relaxing, dread, irritability, difficulty controlling worry, feeling on edge — with only restlessness reliably visible from outside. A transcript rating is therefore almost entirely a judgement about self-description, which makes the wording of your anchors the whole ballgame. Write down what counts as evidence for each item before anyone codes. Our guide to the GAD-7's items, cutoffs and transcript coding covers the anchors, and operationalising a rating scale as a codebook covers how to make the mapping explicit. If two people are rating the same transcripts, compute agreement with the inter-rater reliability calculator.
The original validation reported excellent internal consistency (Cronbach's alpha of 0.92) and good test-retest reliability (intraclass correlation of 0.83). At the cut-point of 10, sensitivity was 89% and specificity 82% for generalised anxiety disorder.
Factor analysis in that study supported the GAD-7 and PHQ-9 measuring distinct though correlated constructs — which is the empirical justification for administering both rather than treating one as a proxy for general distress.
Like the PHQ-9, the GAD-7 was developed with a Pfizer grant and requires no permission to reproduce, translate or distribute. It is hosted in the Tagaroo Scale Library for that reason.
This is a scoring aid for education and research support. It does not diagnose and does not replace clinical assessment. Everything is computed in your browser; no responses are transmitted or stored.
The original validation by Spitzer, Kroenke, Williams and Löwe (2006) reported excellent internal consistency (Cronbach's alpha of 0.92) and good test-retest reliability (intraclass correlation of 0.83). At the cut-point of 10 or above, sensitivity was 89% and specificity 82% for generalised anxiety disorder. Convergent validity is supported by strong correlations with other anxiety measures, and factor analysis supports the GAD-7 and PHQ-9 measuring distinct though correlated constructs.
Each of the seven items is rated 0 (not at all), 1 (several days), 2 (more than half the days) or 3 (nearly every day) over the past two weeks, giving a total from 0 to 21. The conventional bands are 0–4 minimal, 5–9 mild, 10–14 moderate and 15–21 severe anxiety.
A total of 10 or above is the usual threshold for further assessment: in the original validation by Spitzer and colleagues (2006) it gave a sensitivity of 89% and specificity of 82% for generalised anxiety disorder. The GAD-7 also screens reasonably for panic, social anxiety and PTSD, so a positive result indicates that an anxiety disorder is likely somewhere in that family, not that GAD specifically is present.
Depression and anxiety co-occur often enough that screening for one without the other misses a large share of cases, and both instruments use the same two-week window and the same 0–3 response format, so they can be administered as a single short questionnaire. The combined PHQ-4 exists as an ultra-brief version of exactly this pairing.
Yes. Like the PHQ-9 it was developed by Spitzer, Williams, Kroenke and colleagues with a Pfizer grant, and requires no permission to reproduce, translate or distribute.
Written by Enrique Gutiérrez, PhD (Computer Science) — founder of Tagaroo and Associate Professor of Computer Science, working on inter-rater reliability, measurement and annotation methodology (ORCID).
How this page is checked. Scoring rules, item ranges and thresholds are transcribed from the instrument's cited primary sources and covered by automated tests that reproduce each paper's own worked examples. This page has not been reviewed by a licensed clinician, and it is a scoring aid for education and research support rather than a clinical decision tool.
Last verified: 29 July 2026. Formulas, thresholds and cited figures on this page were checked against their original sources on that date. Every calculation runs in your browser; nothing you enter is transmitted or stored.
Tagaroo lets you score the GAD-7 and other instruments against a transcript, with each item rating linked to the words that support it — auditable ratings instead of remembered impressions.