The PHQ-9 and GAD-7: Brief Screening Measures for Depression and Anxiety

The Patient Health Questionnaire-9 (PHQ-9) and the Generalized Anxiety Disorder 7-item scale (GAD-7) were both developed by Drs. Robert Spitzer, Kurt Kroenke, and Janet Williams as part of the larger Patient Health Questionnaire (PHQ) system, a self-administered version of the PRIME-MD diagnostic instrument designed for use in primary care settings.

The PHQ-9 was published in 2001 (Kroenke, Spitzer, & Williams, 2001, Journal of General Internal Medicine). Its nine items correspond directly to the nine DSM-5 diagnostic criteria for major depressive disorder, making it both a severity measure and a tool that can assist in diagnostic assessment.

The GAD-7 followed in 2006 (Spitzer, Kroenke, Williams, & Lowe, 2006, Archives of Internal Medicine). Although named for generalized anxiety disorder, it has demonstrated good sensitivity across multiple anxiety disorders, including panic disorder, social anxiety disorder, and PTSD, making it a useful general anxiety screener.

Both instruments are in the public domain and require no licensing fees, which has contributed significantly to their widespread adoption. They are among the most cited screening instruments in the psychiatric literature.

PHQ-9: Internal consistency is excellent (Cronbach's alpha 0.86-0.89). Test-retest reliability is strong (intraclass correlation 0.84). At a cutoff score of 10, the PHQ-9 has a sensitivity of 88% and specificity of 88% for major depressive disorder (Kroenke et al., 2001). It performs comparably to clinician-administered instruments like the Hamilton Depression Rating Scale.

GAD-7: Internal consistency is excellent (Cronbach's alpha 0.92). At a cutoff score of 10, the GAD-7 has a sensitivity of 89% and specificity of 82% for generalized anxiety disorder (Spitzer et al., 2006). It also demonstrates good operating characteristics as a screener for panic disorder (sensitivity 74%), social anxiety disorder (sensitivity 72%), and PTSD (sensitivity 66%).

Both instruments use the same four-point Likert scale for the past two weeks: Not at all (0), Several days (1), More than half the days (2), Nearly every day (3).

PHQ-9 scoring (range 0-27): 0-4 Minimal, 5-9 Mild, 10-14 Moderate, 15-19 Moderately Severe, 20-27 Severe. A score of 10 or above is the standard threshold for clinically significant depression.

GAD-7 scoring (range 0-21): 0-4 Minimal, 5-9 Mild, 10-14 Moderate, 15-21 Severe. A score of 10 or above is the standard threshold for clinically significant anxiety.

These are screening instruments, not diagnostic tools. A score above threshold indicates the need for further clinical evaluation but does not constitute a diagnosis. Conversely, a score below threshold does not rule out a disorder.

Item 9 of the PHQ-9 asks about thoughts of self-harm or being better off dead. Any endorsement of this item (a score of 1, 2, or 3) should trigger a follow-up safety assessment, regardless of the total score. This is standard clinical practice across all treatment settings.

The presence of suicidal ideation on Item 9 does not by itself indicate imminent risk, but it requires further evaluation using a structured risk assessment protocol. If you or someone you know is in crisis, contact the 988 Suicide and Crisis Lifeline by calling or texting 988.

Initial screening: Both instruments are routinely administered at intake to establish a baseline severity level. Many integrated care settings administer the PHQ-9 and GAD-7 together as a combined depression and anxiety screen.

Treatment monitoring: Serial administration at regular intervals (every session, biweekly, or monthly) allows clinicians to track symptom trajectory. A change of 5 or more points on the PHQ-9 is generally considered a clinically meaningful change (reliable change index). The same threshold is commonly applied to the GAD-7.

Outcome measurement: Both instruments are widely used as primary outcome measures in clinical trials and quality improvement programs. Their brevity, public domain status, and strong psychometrics make them among the most practical tools available in clinical practice.

Measurement-based care: The systematic use of these instruments to guide treatment decisions (dose adjustments, modality changes, referrals) is a growing standard of care. The STAR*D trial and subsequent research have demonstrated that measurement-based care produces better outcomes than treatment as usual.

0-4 (Minimal): Symptoms are absent or subclinical. No treatment is typically indicated based on the score alone. In monitoring contexts, scores in this range after treatment are consistent with remission, which is commonly defined as a PHQ-9 score below 5 (Kroenke et al., 2001).

5-9 (Mild): Symptoms are present but below the clinical threshold. Common responses include watchful waiting, psychoeducation, behavioral activation strategies, and repeating the PHQ-9 at a follow-up visit to assess trajectory. Persistent mild scores across multiple administrations warrant a fuller clinical evaluation.

10-14 (Moderate): This range crosses the standard clinical cutoff of 10. A score of 10-14 typically prompts a structured diagnostic interview and a collaborative treatment plan, which may include psychotherapy, medication, or both, depending on history, duration, functional impairment, and patient preference.

15-19 (Moderately severe): Active treatment is generally indicated. Evidence-based psychotherapy, antidepressant medication, or their combination are first-line considerations, and closer monitoring intervals are common in this range.

20-27 (Severe): Scores in this range indicate severe symptom burden. Combined treatment and expedited follow-up are typical, and the clinician should directly assess safety, functioning, and the need for a higher level of care.

A frequently asked question is what a specific score such as 14 means. A PHQ-9 score of 14 falls in the moderate band, one point below moderately severe. The number itself does not constitute a diagnosis; it indicates symptom severity over the past two weeks and should be interpreted alongside a clinical interview, duration criteria, and functional impairment.

0-4 (Minimal): Anxiety symptoms are absent or negligible. No intervention is typically indicated on the basis of the score.

5-9 (Mild): Symptoms are noticeable but subthreshold. Psychoeducation, self-guided skills practice (controlled breathing, grounding, worry scheduling), and monitoring with repeat administration are common responses.

10-14 (Moderate): A score of 10 or above is the standard clinical cutoff and the level at which further diagnostic evaluation is recommended (Spitzer et al., 2006). Because the GAD-7 also screens positive in panic disorder, social anxiety disorder, and PTSD, an above-threshold score should prompt differential assessment rather than an automatic GAD diagnosis.

15-21 (Severe): High symptom burden. Active treatment with evidence-based psychotherapy (typically CBT), medication, or both is generally indicated, with closer follow-up intervals.

Is a GAD-7 score of 10 high? A score of 10 sits exactly at the validated clinical threshold: it is the point at which the instrument's sensitivity (89%) and specificity (82%) for generalized anxiety disorder were established, and it signals that a clinical conversation is warranted.

How often should the PHQ-9 and GAD-7 be re-administered? The items reference the past two weeks, so administration more often than every two weeks measures overlapping windows. Common cadences are every two weeks during active treatment and every four weeks in maintenance, with a five-point change generally treated as clinically meaningful on either measure.

Can the PHQ-9 and GAD-7 be used together? Yes. They were designed in the same instrument family, share the same response scale, and are routinely co-administered as a combined depression and anxiety screen. Depression and anxiety disorders are highly comorbid, and the two scores together give a more complete severity picture than either alone.

Do high scores equal a diagnosis? No. Both instruments are severity screeners. Diagnosis requires a clinical interview establishing symptom count, duration, and functional impairment, and ruling out medical and substance-related causes. Conversely, a below-threshold score in a person who reports significant distress should not end the conversation.

What about a score of zero with visible distress? Discrepancies between self-report and presentation are clinically informative. Minimization, alexithymia, literacy or language barriers, and impression management can all suppress scores. The instrument supplements clinical judgment; it never replaces it.