What is the GAD-2 and how is it scored?

The GAD-2 is the first two questions of the GAD-7, each rated 0 to 3 over the past two weeks, for a total of 0 to 6. In the Plummer and colleagues (2016) review, a score of 3 or higher was the usual cut point for a likely generalized anxiety disorder.

Should a clinic use the GAD-2 or the GAD-7 to screen for anxiety?

Many clinics use both. The GAD-2 is quick for a first pass, and its brevity helps completion. A positive GAD-2 is then usually followed by the full GAD-7, which the review found slightly more accurate and which gives a severity score you can track over time.

Does a positive GAD-2 mean a patient has anxiety?

No. The GAD-2 screens for a likely disorder; it does not diagnose one. In the meta-analysis it missed about one in four real cases and flagged some people who did not have the disorder, so a score points your team to look closer rather than settling the question.

Maintenance

Is the GAD-2 as accurate as the GAD-7 for anxiety?

A person sits on a bench writing, looking up a path toward a house on a hill.

Last updated November 7, 2026

Short answer

Close, but not quite. In a diagnostic meta-analysis, the 2-item GAD-2 flagged generalized anxiety about 76% of the time and cleared people without it about 81% of the time. The 7-item GAD-7 did a little better. The GAD-2 is a fast first filter, not a final answer.

The research

Plummer, F., Manea, L., Trepel, D., & McMillan, D. (2016). Screening for anxiety disorders with the GAD-7 and GAD-2: a systematic review and diagnostic meta-analysis. General Hospital Psychiatry, volume 39, pages 24 to 31.

Sample The review pooled the published studies that tested each questionnaire against a diagnostic interview, the recognized yardstick. Fewer studies had tested the 2-item GAD-2, so its estimate rests on a smaller base than the GAD-7.

Published 2016

Journal General Hospital Psychiatry

Faye Plummer, Laura Manea, Dominic Trepel, and Dean McMillan pooled the studies that checked two short anxiety questionnaires against a diagnostic interview, which served as the yardstick. The GAD-7 is a seven-item form a person fills out about anxiety over the past two weeks. The GAD-2 is just its first two items. At a GAD-7 score of 8 or higher, the pooled sensitivity was 0.83 and the specificity was 0.84, which means the form caught about 83% of the people who had generalized anxiety and correctly cleared about 84% of those who did not. The two-item GAD-2, at a score of 3 or higher, reached a pooled sensitivity of 0.76 and a specificity of 0.81, so it caught about 76% of real cases and cleared about 81% of people without the disorder. The confidence ranges around the GAD-2 numbers were wide, because fewer studies had tested it, so that estimate is less settled than the one for the GAD-7. The authors concluded that both scales are acceptable screens for generalized anxiety, and that a slightly lower GAD-7 cut point of 8 can be used to catch more cases.

Read the paper

What does this mean for your clinic?

Your clinic lives on short questionnaires. The shorter the form, the more often a patient actually finishes it, and the easier it is to drop into a busy visit or a between-visit check-in. So the two-question GAD-2 is tempting. The honest question is what you give up by cutting five questions, and this review puts a number on it.

The answer is: a little, but not nothing. Measured against a diagnostic interview, the GAD-2 caught about 76% of people who had generalized anxiety and cleared about 81% of those who did not. The full GAD-7 was a step better, at about 83% and 84%. In plain terms, the GAD-2 misses roughly one in four real cases, while the GAD-7 misses closer to one in six. Both will also raise some false alarms. Neither one diagnoses anything on its own.

So the two forms do different jobs. The GAD-2 is a fast first filter: use it to decide who gets a closer look, and expect that a low score is not a clean bill of health. The GAD-7 is the better tool when you want a severity number you can act on and track over time, because its five extra questions pin down how much a person is struggling and give you more room to see a score move between visits. A reasonable pattern is to open with the GAD-2 and follow a positive screen with the full GAD-7.

Now the limits, because they matter here. The GAD-2 estimate rests on fewer studies, so its true accuracy is less certain than the single headline number suggests. These figures come from research settings measured against a trained interview, and your own population may read a little differently. And a shorter screen trades away detail: it can tell you something may be wrong, but not how bad, and not whether it is anxiety rather than something that looks like it. Treat either score as a prompt for a clinician to confirm, not as the last word.

The takeaway

The 2-item GAD-2 caught about 76% of anxiety cases against a diagnostic interview, a bit under the GAD-7's 83%, so it is a quick filter rather than a replacement.
A workable pattern is to screen with the GAD-2 and follow a positive result with the full GAD-7, which gives a severity score you can track between visits.
Fewer studies have tested the GAD-2, so read its accuracy as a careful estimate, and have a clinician confirm any score before it changes a plan.

Terms in this page

Between-session support Mental readiness

This is the work Asendar™ was built for: collecting the same validated anxiety screens between visits and keeping the scores in front of your care team, so a short first filter and the fuller follow-up both land where someone can act on them, not weeks later at the next appointment.

Start a 15-day free trial for your clinic

Track CoCM minutes, catch patients at risk of missing the code, export a billing packet. No card to start, $199/mo per clinician after the trial.

Explore more in the Maintenance pillar