A seven-question quiz that takes two minutes — can it really tell you anything meaningful about your anxiety? It's a fair skepticism. Short self-report tools can feel too simple to trust. So it's worth looking at what the actual research says about how well the GAD-7 performs.
The short answer: for what it's designed to do — screen for generalized anxiety — the GAD-7 is well-validated and performs strongly. But "accurate" means something specific for a screening tool, and understanding that is the key to using your score wisely. If you want your own number first, our free GAD-7 test takes about two minutes.
Where the Numbers Come From
The GAD-7 was developed by Spitzer, Kroenke, Williams, and Löwe and published in 2006, in a study of more than 2,700 patients across 15 primary care clinics. Crucially, the researchers didn't just write seven questions and call it done — they validated the scale against assessments by mental health professionals. That comparison is what lets us talk about its accuracy in concrete terms.
Two measures matter most for any screener: sensitivity and specificity.
Sensitivity and Specificity
These two words sound technical but the ideas are simple.
Sensitivity is how good the test is at catching people who do have the condition — its ability to avoid missing cases. Specificity is how good it is at correctly clearing people who don't have it — its ability to avoid false alarms.
In the validation research, at the standard cut-point of 10 or higher, the GAD-7 showed:
- Sensitivity of about 89% — it catches roughly 9 out of 10 people who have generalized anxiety disorder.
- Specificity of about 82% — it correctly clears roughly 8 out of 10 people who don't.
For a two-minute, seven-item questionnaire, those are genuinely strong numbers. They're why the GAD-7 became one of the most widely used anxiety measures in the world — in primary care, mental health clinics, and research alike.
| Measure | At cut-point ≥10 | What it means |
|---|---|---|
| Sensitivity | ~89% | Catches ~89% of true cases |
| Specificity | ~82% | Correctly clears ~82% of non-cases |
The cut-point of 10 was chosen specifically because it balances these two well. Lower the threshold and you catch more cases but raise false alarms; raise it and you get fewer false alarms but miss more people. Ten is the sweet spot the data points to. (GAD-7 score interpretation covers all the ranges and thresholds.)
Reliability: Does It Give Consistent Results?
Accuracy isn't only about catching the right people — it's also about consistency. A good test should give stable, repeatable results. The GAD-7 performs well here too:
Internal consistency — whether the seven items hang together as a coherent measure — is excellent, with a Cronbach's alpha around 0.92. In plain terms, the questions reliably measure the same underlying thing rather than pulling in different directions. (Each item targets a distinct facet of anxiety, as the GAD-7 questions explained walks through.)
Test-retest reliability — whether you get a similar score taking it twice in a stable period — is also strong. That's what makes the GAD-7 useful for tracking change over time: if your score moves, it's more likely reflecting a real change than random noise.
What "Accurate" Does Not Mean
Here's the part that matters most. The GAD-7 is an accurate screening tool — and a screen is not a diagnosis. Keeping that distinction clear prevents the two most common ways people misread their score.
A high score is not a diagnosis. Even with 82% specificity, some people who score 10+ won't actually have generalized anxiety disorder — that's what the 18% of false positives means. A high score is a strong signal to look further, not a conclusion. Diagnosis requires a clinical assessment that weighs your full history, the duration of symptoms, and whether they're better explained by something else.
A low score is not a guarantee. Sensitivity of 89% is excellent, but it isn't 100% — roughly 1 in 10 true cases is missed. And the scale is built for generalized anxiety specifically, so it can run low even when another pattern is present. If your anxiety comes in discrete spikes, panic attack vs. anxiety attack is relevant; if it's mainly social, see social anxiety vs. generalized anxiety. For more on why a normal number isn't always the full story, see what is a normal GAD-7 score.
The Limits Worth Knowing
Beyond the screen-vs-diagnosis point, a few honest limitations:
It's self-report. The GAD-7 measures how you perceive and report your symptoms. People who minimize, or who've normalized a high baseline through high-functioning anxiety, can score lower than their experience warrants. The number is only as accurate as the honesty behind it.
It's a two-week snapshot. Accuracy is anchored to the recent fortnight. An unusually calm or unusually rough two weeks can pull the score away from your baseline. The trend across several administrations is more informative than any single result.
It doesn't identify causes. Anxiety-like symptoms can come from thyroid problems, certain medications, or even too much caffeine. A high score tells you symptoms are present, not why.
How It Compares to Other Anxiety Measures
The GAD-7 isn't the only anxiety scale, and seeing where it sits helps explain why it's so widely used. Older instruments like the Hamilton Anxiety Rating Scale (HAM-A) and the Beck Anxiety Inventory (BAI) are well-respected, but they were built for different jobs. The HAM-A is clinician-administered — it needs a trained professional to score it — which makes it impractical for quick self-screening. The BAI is self-report but leans heavily on physical symptoms of anxiety, which can overlap with panic and other conditions.
The GAD-7's advantage is fit-for-purpose design: it's short enough to complete in two minutes, validated specifically against generalized anxiety, and balanced across cognitive, physical, and emotional symptoms rather than weighted toward any one. That combination — brief, self-administered, well-validated, and freely available — is why it became a default in primary care and research, even though it isn't necessarily "more accurate" than longer tools in every situation. For most people wanting a quick, trustworthy read on generalized anxiety, it hits the sweet spot.
So — Should You Trust It?
Yes, with the right framing. The GAD-7 is a well-validated, reliable screening instrument with strong sensitivity and specificity for generalized anxiety. Used as intended — as a quick, repeatable check that flags whether anxiety symptoms warrant a closer look — it's about as good as a short self-report tool gets.
What it can't do is replace a clinician. The most accurate way to use your score is as the start of a conversation: a concrete, evidence-based number you can bring to a doctor or therapist, not a verdict you reach on your own.
And whatever your score: if you're having thoughts of self-harm or feeling like you can't go on, please reach out now. In the US, the Suicide & Crisis Lifeline is available 24/7 by call or text at 988.
The Bottom Line
The GAD-7 is accurate at the job it was built for — screening — with roughly 89% sensitivity and 82% specificity at the cut-point of 10, plus strong reliability. It's not a diagnostic test, and it's specific to generalized anxiety, so a single score should be read as a well-founded signal rather than the final word.
Want your own evidence-based number? Take our free GAD-7 test — two minutes, no signup, with a full interpretation of what your score means.