The GAD-7 is scored by adding up the point values for all seven items, giving you a total between 0 and 21. Each item is rated on a four-point scale: 0 for “not at all,” 1 for “several days,” 2 for “more than half the days,” and 3 for “nearly every day.” The higher the total, the more severe the anxiety symptoms.
The Seven Items and Point Values
The questionnaire asks how often, over the past two weeks, you have been bothered by each of seven problems:
- Feeling nervous, anxious, or on edge
- Not being able to stop or control worrying
- Worrying too much about different things
- Trouble relaxing
- Being so restless that it’s hard to sit still
- Becoming easily annoyed or irritable
- Feeling afraid as if something awful might happen
For each item, the person selects one of four responses. “Not at all” scores 0 points. “Several days” scores 1. “More than half the days” scores 2. “Nearly every day” scores 3. You simply add all seven item scores together for the total. If someone answered “several days” on three items and “not at all” on the remaining four, their total would be 3.
What the Total Score Means
The total falls into one of four severity ranges:
- 0 to 4: Minimal anxiety
- 5 to 9: Mild anxiety
- 10 to 14: Moderate anxiety
- 15 to 21: Severe anxiety
A score of 10 or higher is the most widely used threshold for flagging clinically significant anxiety. At that cutoff, the GAD-7 correctly identifies about 89% of people who have generalized anxiety disorder (sensitivity) and correctly rules it out in about 82% of people who don’t (specificity). Some research suggests that lowering the cutoff to 8 or 9 can catch a few more true cases without dramatically increasing false positives, so clinicians occasionally adjust the threshold depending on the setting.
Tracking Changes Over Time
The GAD-7 is often given repeatedly to monitor whether symptoms are improving, staying the same, or getting worse. When comparing scores from one visit to the next, a drop of about 4 points is considered the smallest change that reflects a meaningful real-world difference in how someone feels. A shift from 16 to 12, for example, would cross that threshold and suggest a genuine improvement rather than normal score fluctuation. Smaller changes may just reflect day-to-day variation in symptoms.
What a Score Does and Doesn’t Tell You
The GAD-7 is a screening tool, not a diagnostic test. A high score signals that anxiety symptoms are present and worth evaluating further, but it doesn’t confirm a diagnosis of generalized anxiety disorder or any other anxiety condition on its own. A formal diagnosis requires a clinical interview that considers the full picture: how long symptoms have lasted, whether another condition better explains them, and how much they interfere with daily life.
False positives are a real consideration. Some people score above 10 without having a diagnosable anxiety disorder, because the questions capture stress responses that can be temporary or situational. That’s why the score is best used as a starting point for a conversation rather than a final answer. On the flip side, someone scoring in the mild range (5 to 9) may still benefit from support depending on their circumstances, so a low score doesn’t automatically mean everything is fine.
Tips for Accurate Scoring
The two-week lookback period matters. Responses should reflect the past 14 days specifically, not how someone felt months ago or on one particularly bad day. If you’re administering the questionnaire, make sure the person understands that timeframe before they begin. Every item needs a response for the total to be valid. If one or more items are left blank, the score will underrepresent the person’s actual symptom level, and there’s no standard rule for imputing missing answers on a seven-item scale this short.
The GAD-7 also includes an optional follow-up question asking how difficult the reported problems have made it to do work, take care of things at home, or get along with other people. This item isn’t part of the 0 to 21 score, but it adds useful context about functional impact that the numerical total alone doesn’t capture.