What Does My Anxiety Test Score Actually Mean? (GAD-7 Result Guide)
In crisis? Call or text 988 — Suicide & Crisis Lifeline, free and 24/7.
You took the test. You have a number. Now what.
If you've just scored yourself on the GAD-7 (or any anxiety screener) and you're sitting with a result somewhere between 0 and 21, the question that comes next is almost never what was the number. It's what does the number mean for me, this week, this month.
This post is the interpretation layer. For each severity band — minimal, mild, moderate, severe — we cover what the score clinically indicates, what a doctor's default recommendation usually looks like at that level, what's actually evidence-based for self-management, and what red flags would push the conversation from "monitor" to "see someone soon."
The screener itself, if you haven't taken it yet, is at /tools/gad-7 — two minutes, no signup, no data stored. For the deeper context on the GAD-7 as an instrument (history, validation, what it can and can't tell you), see Anxiety: When to See a Doctor — that post is the decision tree; this one is what the result means once you have it.
Quick reference: the four bands
| Score | Band | What it means | Typical next step | |---|---|---|---| | 0-4 | Minimal | Anxiety symptoms below clinical threshold | Self-monitor; re-screen if life changes | | 5-9 | Mild | Real but sub-clinical symptoms | Watch-and-wait; revisit in 2-4 weeks | | 10-14 | Moderate | Clinical attention warranted | Talk to a doctor | | 15-21 | Severe | Clinical attention strongly recommended | Talk to a doctor soon |
The standard clinical cutoff is 10. That's the threshold where most primary care doctors will start a real conversation about anxiety treatment rather than note it and move on. It's not a diagnosis (the GAD-7 doesn't diagnose anything — it screens), but it's the number that changes the next sentence in a clinical visit.
Score 0-4: Minimal anxiety
What this clinically indicates: Whatever anxiety you're experiencing right now is below the threshold where the GAD-7 detects clinical levels. Most adults occasionally score in this band — anxiety in the 0-4 range is part of normal human experience, especially during stressful weeks.
Doctor-recommendation default: None. A 0-4 score in primary care typically gets noted in the chart and the conversation moves on. If you've been screened during a routine visit and scored here, you'll get a check-in next year.
What's evidence-based for self-management: Generic mental-health basics — sleep regularization, ~3-5 sessions of aerobic exercise per week, caffeine moderation, an emotional vocabulary that lets you name what you're feeling. None of this is anxiety-specific because there isn't really an anxiety problem to specifically address.
When to escalate: If a 0-4 score is suspiciously low given what you know about your life right now — high stress, sleep loss, a recent loss or major change — the score may be reflecting suppression rather than absence. People who are good at not feeling their feelings can produce a misleadingly low GAD-7. Re-take in 4 weeks. If the score climbs, that's the real signal.
Caveat: A 0-4 doesn't mean nothing is wrong. The GAD-7 specifically measures generalized anxiety symptoms (worry, restlessness, irritability, etc.) over the past two weeks. It does NOT measure depression, PTSD, social anxiety, panic disorder, OCD, or any of the dozens of other things that can hurt. If you're feeling like something is off and your GAD-7 came back clean, consider the PHQ-9 depression screener (a different instrument for a different set of symptoms) or talk to a doctor anyway.
Score 5-9: Mild anxiety
What this clinically indicates: Real anxiety symptoms, sub-clinical threshold. Most adults will be in this band at some point — stressful weeks, life transitions, major decisions, grief. A 5-9 is not nothing, but it's not yet at the level where clinical intervention shows efficacy advantage over good self-care.
Doctor-recommendation default: "Watch and wait." A primary care doctor who sees a 5-9 will typically suggest revisiting in 2-4 weeks rather than immediately starting treatment. The reasoning: most 5-9 scores resolve on their own as the trigger passes, and starting treatment too early risks treating noise.
What's evidence-based for self-management:
- Sleep regularization — consistent sleep window with a 30-minute wind-down. Anxiety and sleep loss feed each other; fixing the sleep half often drops the GAD-7 by 1-3 points.
- Aerobic exercise — 20-30 minutes, 3-5x/week, moderate intensity. Multiple meta-analyses show effect sizes comparable to first-line anxiety medications for mild-to-moderate cases (Stonerock et al., 2015 is the cleanest reference).
- Alcohol audit — two weeks alcohol-free is the cleanest natural experiment for whether alcohol is amplifying your baseline. A lot of 5-9 scores drop to 2-4 with this alone.
- Caffeine cap — more than ~400mg/day reliably amplifies anxiety in people predisposed to it. Cut to 200mg for two weeks; if the GAD-7 drops, you've found a lever.
- Identifying triggers — keep a brief note for two weeks on what was happening when the anxiety spiked. Patterns emerge that you can't see from inside the experience.
When to escalate: Two scenarios. First, if 4 weeks of self-care doesn't move the GAD-7 score downward, that's the signal to book a doctor visit even though you're still technically below threshold. Second, if specific symptoms are causing real-world impact — missing work, avoiding social situations, sleep that won't recover — the band-level number matters less than what's actually happening in your life.
Honest framing: A 5-9 is sometimes a "you're going through something hard" score and sometimes a "this is becoming chronic" score. The way to tell the difference is by tracking it. One 5-9 score tells you almost nothing. Three 5-9 scores three weeks apart tell you everything.
Score 10-14: Moderate anxiety
What this clinically indicates: You've crossed the standard clinical cutoff. The GAD-7's sensitivity at 10+ is 89% (per Spitzer et al., 2006), meaning if you actually have a generalized anxiety disorder, the screener will catch it nearly nine times out of ten. The specificity is 82%, meaning some 10+ scores are false positives. Either way, a 10 is the number that gets a real clinical conversation started.
Doctor-recommendation default: Most primary care doctors will offer some combination of:
- A therapy referral — typically CBT-trained, often via the doctor's network or insurance directory. Wait times vary wildly by region; this is the slow path.
- A medication conversation — SSRIs (sertraline, escitalopram) and SNRIs (venlafaxine) are first-line. Buspirone and beta blockers come up for specific patterns. Benzos rarely first-line because of dependence risk. (Anxiety Medications: What Your Doctor Might Not Tell You covers the trade-offs in detail.)
- A workup to rule out medical causes — thyroid (TSH), B12, sometimes cardiac (ECG) depending on physical symptoms. This is standard, not over-cautious.
- Follow-up screen in 4-6 weeks — to see if anything moved.
What's evidence-based for self-management: Everything from the mild band still applies but generally won't be sufficient as a sole intervention. At 10+, professional support produces better outcomes faster than self-care alone, on average. The exception is people who have a clear trigger (acute grief, major life event) where the moderate score is expected to be transient — those people sometimes do recover with self-care plus time.
When to escalate: Multiple paths.
- Score climbs over 4 weeks — book sooner, not later.
- Score is accompanied by depression symptoms — about half of moderate-anxiety scorers also have a moderate or higher PHQ-9 depression score. The combination is more disruptive than either alone and benefits from being addressed together.
- Specific patterns — panic attacks, agoraphobia (avoiding leaving home), severe social anxiety, intrusive thoughts. The GAD-7 doesn't measure these specifically — a doctor visit can identify whether a different screener (Liebowitz Social Anxiety Scale, PCL-5 for trauma, Y-BOCS for OCD) is more appropriate.
- Functional impairment — if you've quietly stopped doing things you used to do (driving, attending events, opening mail, scheduling appointments) because the anxiety has made them feel impossible, the avoidance pattern matters more than the score.
Honest framing: A 10-14 is the band where the cost-benefit math shifts from "self-care first, intervention if needed" to "intervention first, self-care as adjunct." That doesn't mean you have to take medication, doesn't mean you have to start therapy this week — but it does mean a professional conversation is the right next move.
Score 15-21: Severe anxiety
What this clinically indicates: Severe anxiety symptoms. Most people scoring in this band describe their day as substantially shaped by the anxiety: difficulty concentrating, persistent physical symptoms, sleep heavily impacted, multiple areas of life (work, relationships, basic functioning) compromised.
Doctor-recommendation default: Same intervention options as the moderate band, but the conversation moves faster. A doctor seeing a 15+ score is generally more aggressive about offering medication earlier alongside therapy referral, because the symptom load is substantially affecting quality of life and the wait for therapy alone (often weeks) is a real cost.
In some practices, a 15+ score plus specific symptom patterns triggers a referral to psychiatry rather than waiting on a primary care + therapy combination.
What's evidence-based for self-management: Self-care strategies remain helpful — sleep, exercise, alcohol and caffeine management, social support — but at this score level, the cost of trying to do this alone tends to be high. Recovery is faster and more reliable with professional support included.
When to escalate: Faster than the moderate band. The general rule:
- 15-17 → book the appointment this week
- 18-21 → book the appointment today; if your GP can't see you within a week, ask about a telehealth alternative
- Any score in this band accompanied by thoughts of self-harm or feeling that you can't function → this is no longer about the GAD-7 score, it's an urgent visit. Same-day visit to primary care, or call 988 (US Suicide & Crisis Lifeline) if you're in active crisis.
Honest framing: A 15+ is where some people start to feel embarrassed about their score, or to compare themselves to internal averages, or to wonder if they're being dramatic. They're not. The GAD-7 is a measurement, not a moral judgment. A 15 just tells you that the symptoms are at a level where most clinicians have seen the same pattern thousands of times and have a treatment plan that helps.
What if my score doesn't match how I feel?
Three patterns are worth knowing about:
High score, "I don't feel that bad": Some people normalize chronic anxiety to the point where they don't realize what they're scoring is high. If your friends and family describe you as a worrier, if you say "I just have a lot going on" every time someone asks how you are, if you wake up at 3am thinking about tomorrow's email — your felt experience may be calibrated to chronic anxiety. A 15 might feel like "everyone deals with this," but it isn't.
Low score, "I feel terrible": The GAD-7 specifically measures generalized anxiety. It misses other things. If you're scoring 4 but feeling awful, consider whether what you're experiencing is:
- Depression rather than anxiety → PHQ-9
- Burnout rather than anxiety → there's overlap but burnout has a different shape (specifically: cynicism + reduced sense of accomplishment + emotional exhaustion in work context)
- A specific anxiety subtype (social, panic, OCD, trauma) → those need different screeners
- Grief → grief overlaps with anxiety symptoms but isn't a disorder; see Toxic Positivity in Grief for the broader frame
- Something physical (thyroid, anemia, postpartum, perimenopause) → worth a medical workup
Score swings week-to-week: Normal. The GAD-7 asks about the past two weeks. A bad two weeks (work crunch, family conflict, a big decision) will produce a higher score. Track 3-4 scores over 6-8 weeks to find your real baseline.
What to do this week, regardless of score
Independent of band:
- Write down the score and the date. Tracking is more useful than a single point.
- Pick ONE evidence-based lever to test. Sleep, exercise, alcohol, caffeine. Two weeks. Re-screen. See what moved.
- If you're considering professional help, start the doctor conversation here — that post walks through the threshold signals, what a visit looks like, and what to bring.
- If your score includes depression symptoms, screen for those too — PHQ-9 is the parallel tool.
A note on what online tests can and can't tell you
The GAD-7 is a screening instrument, not a diagnostic one. A score of 10 doesn't mean you have generalized anxiety disorder. It means you've passed a threshold where, on average, clinical evaluation is worthwhile. The diagnosis requires a licensed clinician working through DSM-5 criteria (which include duration, functional impairment, ruling out other causes, etc.) — not a number from an online quiz, including this one.
The good news is that the screener and the diagnosis usually agree. The bad news is that they don't always, and the gap matters. A high GAD-7 with no actual disorder happens. A low GAD-7 with a real underlying disorder also happens. The way to find out which you are is to talk to someone whose job is to figure that out, not to keep retaking the test.
For more on how reliable these screeners actually are and what the research shows about their accuracy, see How Accurate Are Online Anxiety Tests?.
Sources & further reading
- Spitzer RL, Kroenke K, Williams JBW, Löwe B (2006). "A brief measure for assessing generalized anxiety disorder: the GAD-7." Archives of Internal Medicine — the validation study with the 10-cutoff threshold (89% sensitivity, 82% specificity)
- NIMH — Anxiety Disorders
- Anxiety & Depression Association of America — Symptoms & Treatment
- Stonerock GL et al. (2015). "Exercise as Treatment for Anxiety: Systematic Review and Analysis." Annals of Behavioral Medicine
- 988 Suicide & Crisis Lifeline
Related Reading
- GAD-7 Anxiety Self-Screener: Take it now if you haven't yet.
- Anxiety: When to See a Doctor: The decision tree before you book.
- How Accurate Are Online Anxiety Tests?: What the research actually says about screener reliability.
- PHQ-9 Depression Self-Screener: The parallel tool — anxiety and depression are heavily comorbid.
- Anxiety Medications: What Your Doctor Might Not Tell You: The medication landscape if your score is 10+ and meds come up.
- The 3am Anxiety Action Plan: What to do at 3am when your mind won't stop.
- The Complete Anxiety Guide: The pillar — types, causes, treatments.
ILTY is a mental-health support tool, not a substitute for professional care. If you're in crisis, call or text 988.
Share this article

ILTY Team
AI Mental Health Companion
Building an AI companion that actually helps with your mental health.
Get mental health insights in your inbox
No fluff, no toxic positivity — just what actually helps.
Related Support
ILTY can help with what you're reading about.
Related Articles
Anxiety: When to See a Doctor (A Self-Assessment Decision Guide)
Most people wait too long. The threshold for 'see a doctor about anxiety' is much lower than the threshold most people set for themselves — and the GAD-7 (the 7-question screen clinicians use) is the cleanest way to find out where you actually are. Here's the decision tree, what a doctor visit actually looks like, and what to bring.
How to Help Someone Having a Panic Attack
Someone you care about is panicking, and you have no idea what to do. Here's exactly how to help — and what to avoid saying.
How to Stop a Panic Attack in Under 5 Minutes
Your heart is pounding, you can't breathe, and you're convinced something is seriously wrong. Here's exactly what to do right now.