GAD-7: Scoring, Cutoffs & Anxiety Screening
GAD-7 guide — 7-item structure, scoring table with 5/10/15 cutoffs, 2006 validation psychometrics, public-domain licensing, and EHR workflow.
GAD-7: Scoring, Cutoffs, and Anxiety Screening Workflow
GAD-7 guide — 7-item structure, scoring table with 5/10/15 cutoffs, validation psychometrics, public-domain licensing, and facility workflow.
How to use this page
Use this page to understand GAD-7 scoring and to plan anxiety-screening documentation workflows. A GAD-7 score supports — but never replaces — clinician judgment; keep patient-specific decisions in your clinical and payer workflows.
Last source check: June 12, 2026
Plan the measurement workflow
Use these sections to plan ownership, documentation, and follow-up around validated measure materials and your program's care workflows.
Define The Monitoring Use Case
- Document where anxiety screening or monitoring fits in the organization's intake, follow-up, or quality workflow.
- Keep source access, attribution, and version details connected to any local documentation prompts.
Connect Results To Review
- Route result review into progress-note and treatment-plan update workflows where clinically relevant.
- Set recurrence timing through program policy and clinical ownership.
Route Decisions Through Clinical Review
- Configure measure wording, score context, documentation handoffs, and follow-up steps around validated source materials.
- Keep diagnosis, placement, medical-necessity, and coverage decisions in the team's clinical and payer workflows.
What the GAD-7 Is
The Generalized Anxiety Disorder-7 (GAD-7) is a seven-item, self-report measure of anxiety symptom severity. The patient rates how often each symptom has bothered them over the past two weeks on a 0–3 frequency scale (0 = not at all, 1 = several days, 2 = more than half the days, 3 = nearly every day), for a total score of 0–21.
The GAD-7 was developed by Robert Spitzer, Kurt Kroenke, Janet Williams, and Bernd Löwe, with development funded by an educational grant from Pfizer, and validated in a 2006 study in Archives of Internal Medicine ↗. Like the rest of the PHQ Screeners family, it is public domain — PHQ Screeners ↗ states the screeners may be reproduced, translated, displayed, or distributed without permission.
What the Seven Items Cover
The table below describes the symptom domain each item addresses. For exact item wording and translations, use the source forms at phqscreeners.com ↗ and keep the attribution that appears on them.
An unscored follow-up question asks how difficult the endorsed problems have made it to work, manage home responsibilities, or get along with others — functional-impact context that belongs in the chart alongside the score.
| Item | Symptom domain |
|---|---|
| 1 | Feeling nervous, anxious, or on edge |
| 2 | Not being able to stop or control worrying |
| 3 | Worrying too much about different things |
| 4 | Trouble relaxing |
| 5 | Restlessness — finding it hard to sit still |
| 6 | Becoming easily annoyed or irritable |
| 7 | Feeling afraid that something awful might happen |
GAD-7 Scoring Table and Cutoffs
Add the seven item scores for a total of 0–21, then interpret against the published severity bands:
* Response pathways are program- and clinician-determined; the bands orient severity, they do not prescribe treatment.
A total score of 10 or higher is the recommended threshold for further diagnostic evaluation ( Spitzer et al., 2006 ↗ ).
| Total score | Severity band | Common clinical response* |
|---|---|---|
| 0–4 | Minimal | Monitor; rescreen per program policy |
| 5–9 | Mild | Watchful waiting; repeat GAD-7 at follow-up |
| 10–14 | Moderate | Further evaluation; consider anxiety-focused treatment planning |
| 15–21 | Severe | Active treatment indicated per clinical evaluation |
Psychometrics: Why the ≥10 Cutoff
In the 2006 validation study of 2,740 primary-care patients, a GAD-7 score of ≥10 had 89% sensitivity and 82% specificity for generalized anxiety disorder ( Arch Intern Med 2006;166:1092–1097 ↗ ). The same study reported excellent internal consistency and good test-retest reliability, and found that GAD-7 scores correlated as expected with functional impairment — higher scores tracked with more disability days and more clinic visits, which supports using the score as a severity measure rather than only a yes/no screen.
Like its depression-side twin, the GAD-7 is sensitive to change — repeated administrations produce an interpretable trend line, which is what makes it useful for measurement-based care rather than intake-only screening. A score falling across administrations is evidence of symptom improvement in a form payers and accreditation surveyors recognize.
A follow-up analysis found the GAD-7 also performs reasonably well as a general anxiety screen — flagging panic disorder, social anxiety disorder, and PTSD, not only GAD ( Kroenke et al., 2007, Ann Intern Med ↗ ). In practice that means an elevated GAD-7 says “evaluate anxiety,” not “this is GAD.” Programs that need disorder-specific confirmation follow an elevated GAD-7 with a diagnostic interview, and where PTSD is suspected, a trauma-specific measure such as the PCL-5.
GAD-2 First, GAD-7 to Confirm
The GAD-2 consists of the first two GAD-7 items (feeling nervous/on edge and uncontrollable worry), scored the same way for a total of 0–6. Programs that need an ultra-brief first-stage screen administer the GAD-2 and step up to the full GAD-7 when it is positive. The two-stage model parallels the PHQ-2 → PHQ-9 pathway used for depression, and it serves the same workflow purpose: universal two-question screening stays sustainable in high-volume settings, while only positives generate the full instrument and the clinical review it requires.
Anxiety vs. Withdrawal: GAD-7 in Addiction Treatment Settings
The GAD-7 needs the most interpretive care in substance use disorder programs, because anxiety symptoms and withdrawal symptoms overlap almost item for item. Restlessness, irritability, feeling on edge, trouble relaxing, and a sense that something awful is about to happen are core features of alcohol, benzodiazepine, opioid, and stimulant withdrawal — and of post-acute withdrawal in early recovery. A patient three days into detox can produce a severe-range GAD-7 that reflects physiology, not an independent anxiety disorder.
Practical implications for facility workflows:
- Timing matters more than the score. A GAD-7 administered at detox admission establishes a baseline, but the score that informs diagnosis is the one taken after acute withdrawal resolves. Many programs re-administer at the residential or PHP hand-off and again at outpatient step-down, so the trend line separates resolving withdrawal from persisting anxiety.
- Document the context with the score. A note that records “GAD-7 = 17, day 2 of alcohol withdrawal management, CIWA-Ar-monitored” reads very differently in payer review than a bare “GAD-7 = 17.” Pairing the score with withdrawal-scale context keeps the record honest about what is being measured.
- Diagnosis waits for the clinical picture. Substance-induced anxiety symptoms that resolve with stabilization do not support an independent GAD diagnosis. The GAD-7 cannot make that distinction — only longitudinal observation and clinical interview can, which is exactly the boundary this page’s claims framework draws.
- Persisting elevation is a treatment-plan trigger. When scores stay elevated after stabilization, that is the moment to add an anxiety-focused problem, goal, and measurable objective to the treatment plan, with the post-stabilization GAD-7 as the baseline.
Special Populations
- Adolescents. The GAD-7 was validated in adults. Programs serving minors should confirm which anxiety measure their clinical leadership has adopted for adolescents rather than assuming adult cutoffs transfer.
- Older adults. Worry content and somatic presentation differ in older populations, and medical illness can inflate somatic-adjacent items; scores orient severity but warrant medical context.
- Perinatal settings. Anxiety screening in pregnancy and postpartum is increasingly standard alongside depression screening; the GAD-7 is commonly used, with follow-up routed through perinatal-specific protocols.
GAD-7 Across Levels of Care
A facility running multiple levels of care gets the most from the GAD-7 by standardizing when it is administered, not just that it is administered: at intake to every program, at each level-of-care transition, and on the cadence each program’s policy sets for active treatment. That produces a single trend line that follows the client from detox through residential, PHP, IOP, and outpatient care — outcome evidence for continued-stay review at every step, and a clean clinical story at discharge. Route the administration points through the same workflow that owns progress notes and treatment plan updates so no score lands in the chart without a reviewer.
Licensing: Public Domain, Attribution Kept
The GAD-7 requires no license and no permission fee. PHQ Screeners ↗ covers the GAD-7 under the same no-permission reproduction language as the PHQ family. Keep attribution and version details connected when configuring local forms — good measurement hygiene, and it keeps your documentation source-aligned for accreditation review.
That free status contrasts with licensed instruments such as the Beck Depression Inventory (BDI-II), which must be purchased from the rights holder and whose items may never be reproduced.
From Score to Treatment Plan
A GAD-7 score earns its place in the chart when it becomes the measurable baseline in a treatment-plan objective (for example, “reduce GAD-7 from 16 to below 10 within 90 days, measured monthly”) and repeated outcome evidence for payer review of continued care. For anxiety-specific goal and objective language, see our anxiety treatment plan guide, and for the diagnosis-coding side, our ICD-10 codes for anxiety guide and F41.9 code guide.
Connect each administration to the progress note documenting the encounter and to any treatment plan update that follows. The boundary holds: a score is evidence for clinical judgment, never a substitute, and cadence is program policy — not a universal schedule.
Billing the Screening
When the GAD-7 is administered and scored with a documented result, CPT 96127 (brief emotional/behavioral assessment with a standardized instrument) commonly applies — see our 96127 billing guide for units, frequency limits, and payer notes in original wording. Many programs administer the GAD-7 and PHQ-9 together at intake, since anxiety and depression frequently co-occur; for the definition and clinical picture of GAD itself, see our generalized anxiety disorder glossary entry.
EHR supports GAD-7 administration in-app with auto-scoring on submission. Score trending is currently available via reports; visual trend graphs are on our roadmap.
Source guidance
What the public references support
These notes summarize public reference material so behavioral-health teams can verify how an assessment fits their documentation process.
PHQ Screeners includes GAD-7 in the screener family covered by no-permission reproduction language.
Keep attribution, version, and source context connected when configuring local forms or workflows.
GAD-7 can be framed around anxiety severity orientation and clinically reviewed monitoring, but the current source set does not establish a universal recurrence cadence.
Tie review timing to program policy and clinical review.
Published GAD-7 severity bands are 0–4 minimal, 5–9 mild, 10–14 moderate, and 15–21 severe, with a score of 10 or higher the recommended threshold for further evaluation.
Severity bands orient interpretation; diagnosis and treatment decisions remain with the clinician. New claim — pending legal/clinical review.
The 2006 validation study (Spitzer, Kroenke, Williams & Löwe, Arch Intern Med) reported 89% sensitivity and 82% specificity for generalized anxiety disorder at the cutoff of 10 or higher.
Cite the primary study; psychometrics describe study samples, not any individual patient. New claim — pending legal/clinical review.
Measurement-based-care context supports standardized instruments over the course of service.
Use as accreditation and documentation context, not clinical advice.
Outcome-measure programs should document why repeated measurement is used and avoid treating a single instrument schedule as universal.
Use this to keep measurement timing tied to program policy and clinical review.
Frequently Asked Questions
Related workflow
- How is the GAD-7 scored?
- What does a GAD-7 score of 10 mean?
- Is the GAD-7 free to use?
- GAD-7 vs PHQ-9 — what's the difference?
- How can a behavioral health team operationalize GAD-7 for anxiety outcome monitoring?
- How should GAD-7 scores inform clinical and payer decisions?
Continue the measurement path
References
Source material
Use these links to confirm source details and align your local assessment workflow.
Coordinate measurement review
Connect measures to next steps
Review how connects anxiety-measure review, progress notes, treatment-plan updates, reminders, and reporting tasks.
For informational purposes only: This page provides general information for behavioral-health teams evaluating assessment workflows. It is not medical, clinical, diagnostic, legal, billing, accreditation, or compliance advice, and it is not a recommendation to select, administer, score, interpret, or bill any instrument for any specific patient, program, payer, or jurisdiction. Use official measure materials, publisher and license terms, instrument instructions, validated scoring guidance, payer and accreditation requirements, program policies, and qualified clinical, compliance, and legal review. Measure names and trademarks belong to their respective owners; does not supply, reproduce, license, endorse, or replace restricted instruments, item text, scoring keys, official forms, or clinical interpretation unless expressly stated in writing.
Reference tables
| Item | Symptom domain |
|---|---|
| 1 | Feeling nervous, anxious, or on edge |
| 2 | Not being able to stop or control worrying |
| 3 | Worrying too much about different things |
| 4 | Trouble relaxing |
| 5 | Restlessness — finding it hard to sit still |
| 6 | Becoming easily annoyed or irritable |
| 7 | Feeling afraid that something awful might happen |
| Total score | Severity band | Common clinical response* |
|---|---|---|
| 0–4 | Minimal | Monitor; rescreen per program policy |
| 5–9 | Mild | Watchful waiting; repeat GAD-7 at follow-up |
| 10–14 | Moderate | Further evaluation; consider anxiety-focused treatment planning |
| 15–21 | Severe | Active treatment indicated per clinical evaluation |
Common questions
Official sources
- a 2006 study in Archives of Internal Medicinepubmed.ncbi.nlm.nih.gov
- Kroenke et al., 2007, Ann Intern Medpubmed.ncbi.nlm.nih.gov
- Joint Commission measurement-based care FAQjointcommission.org