Home / Resources / Psychological Measures / GAD-7
Generalized Anxiety Disorder 7-Item Scale:
Scoring and Interpretation
The Generalized Anxiety Disorder 7-item scale (GAD-7) is a self-report measure of anxiety symptom severity over the past two weeks, scored from 0 to 21 (Spitzer et al., 2006). The topics are GAD-7 scoring, what the four bands and the cut-off of 10 can and cannot show, SPSS and R code, and a methods paragraph for a manuscript.
What the GAD-7 Measures
The GAD-7 was devised to identify probable generalized anxiety disorder and to quantify how severe a person’s anxiety symptoms have been recently. Its criterion validation ran across 15 US primary-care practices, where questionnaire scores were compared with independent assessments by mental health professionals (Spitzer et al., 2006).
Researchers often use it as a continuous patient-reported anxiety outcome in intervention research, public-health surveys and medical studies. The scale was built around generalized anxiety, but its score may reflect broader anxiety distress. It therefore works as a screening and severity measure and not as a standalone assessment of every diagnostic criterion.
Structure of the GAD-7
- Items: seven symptom items, each answered with reference to the last two weeks.
- Response scale: four-point frequency scale coded 0 to 3.
- Subscales: none. One total from 0 to 21 is the standard output.
- Reverse-scored items: none.
- Unscored content: the functional-difficulty question on some forms is separate from the total.
- Short form: the two-item GAD-2 uses the first two GAD-7 symptom items and has its own 0 to 6 range, so it must be read on its own scale.
The instrument itself is available from PHQ Screeners (phqscreeners.com) and the University of Washington National HIV Curriculum (University of Washington, n.d.).
How to Score the GAD-7
- Confirm that gad1 to gad7 each hold a valid response from 0 to 3.
- Add the seven items to obtain gad7_total, which ranges from 0 to 21. There is no reversal and no transformation.
- Apply a missing-item rule. The original validation sets no universal prorating method, so the code below requires all seven responses. If your study uses a different rule, state it in the methods section.
* Illustrative syntax: variables must already be valid item scores (0 to 3).
COMPUTE gad7_total = SUM.7(gad1, gad2, gad3, gad4, gad5, gad6, gad7).
EXECUTE.# data: one row per respondent; items gad1 to gad7 coded 0 to 3.
data$gad7_total <- rowSums(data[c("gad1", "gad2", "gad3", "gad4", "gad5", "gad6", "gad7")], na.rm = FALSE)Interpreting GAD-7 Scores
| Score band | Interpretation |
|---|---|
| 0 to 4 | Minimal anxiety |
| 5 to 9 | Mild |
| 10 to 14 | Moderate |
| 15 to 21 | Severe |
Source: Spitzer et al. (2006).
A total of 10 or more is the commonly cited screening threshold. In the original US primary-care study it gave sensitivity of 89% and specificity of 82% for generalized anxiety disorder, whereas a later meta-analysis found that thresholds from about 7 to 10 produced similar pooled trade-offs across studies (Plummer et al., 2016). Higher totals reflect more frequent anxiety symptoms, and a positive screen does not establish a disorder without a clinical assessment.
Predictive value shifts with the reference diagnosis, the clinical setting and the prevalence of the condition. A two-week symptom score also cannot establish the six-month duration criterion that applies to generalized anxiety disorder.
Psychometric Properties
| Property | Published finding | Source |
|---|---|---|
| Internal consistency | Cronbach’s α = .92 | Spitzer et al. (2006) |
| Test-retest | Intraclass correlation = .83 | Spitzer et al. (2006) |
| Construct validity | Scores were associated with impairment in all six SF-20 functioning domains | Spitzer et al. (2006) |
| Diagnostic accuracy | At 10 or more: sensitivity .89, specificity .82 in the original study | Spitzer et al. (2006) |
| Cross-study accuracy | Meta-analysis at 8 or more: pooled sensitivity .83 and specificity .84, with heterogeneity present | Plummer et al. (2016) |
These are source-specific results and cannot be assumed to hold as reliability or accuracy estimates in a new sample.
Using the GAD-7 in a Study
The summed GAD-7 usually serves as a continuous outcome or predictor, for example in a Multiple Linear Regression or another generalized linear model, provided the distribution and the residuals support it. Highly skewed totals may call for an alternative model. When the prevalence of elevated symptoms is a secondary endpoint, prespecify the cut-off and report the numerator and denominator, rather than describing every positive screen as a confirmed diagnosis.
Report internal consistency from your own sample through Reliability Analysis (Cronbach’s Alpha), and inspect the distribution for ceiling and floor effects. Studies that compare demographic or language groups may need to examine differential item functioning with Item Response Theory (IRT) or test Measurement Invariance before interpreting differences in means. Sample size depends on the model and the estimand, not on a blanket rule for seven items.
Related field: Psychology & Behavioral Sciences
Reporting the GAD-7
Anxiety symptoms were measured with the Generalized Anxiety Disorder 7-item scale (GAD-7; Spitzer et al., 2006). The seven items were scored from 0 to 3 and summed (range = 0 to 21), with higher totals indicating more severe symptoms. The total served as [specified analysis variable], and internal consistency in this sample was [α = XX].
References
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 metaanalysis. General Hospital Psychiatry, 39, 24-31. https://doi.org/10.1016/j.genhosppsych.2015.11.005
Spitzer, R. L., Kroenke, K., Williams, J. B. W., & Löwe, B. (2006). A brief measure for assessing generalized anxiety disorder: The GAD-7. Archives of Internal Medicine, 166(10), 1092-1097. https://doi.org/10.1001/archinte.166.10.1092
University of Washington. (n.d.). Generalized Anxiety Disorder 7-item (GAD-7). https://www.hiv.uw.edu/page/mental-health-screening/gad-7
Frequently Asked Questions
Does a GAD-7 score of 10 or more mean someone has generalized anxiety disorder?
No. It is a screening threshold, and a positive screen calls for diagnostic follow-up where indicated. In a research report, call it a positive screen, not a case, unless a diagnostic interview confirmed the condition.
Can I use GAD-7 totals in linear regression?
Often, yes, provided the distribution, the model residuals and the interpretation are suitable. Highly skewed totals may be better served by another model, so check the fit before you commit to the linear form.
Can GAD-2 and GAD-7 totals be compared directly?
No. The GAD-2 has a 0 to 6 range and its own validation, so a GAD-2 total cannot be placed on the GAD-7 scale. Record which form was collected for every participant.
How should missing GAD-7 items be handled?
The code here scores only complete responses. If you choose another approach, specify and justify it before analysis and report it in the methods section.
NEXT STEP
Need a GAD-7 Analysis Set Up
or Reviewed?
OnlineSPSS is a statistical consultancy for researchers. Share your design and dataset, and a statistically trained Research Coordinator reviews it. The scope and one fixed project fee are confirmed before work begins.
No Commitment Required
Strictly Confidential
Fixed Project Fee
In-House PhD-Level Statisticians
Same-Day Response
Serving Researchers Since 2014