Home › Tools › PHQ-9 and GAD-7 Scoring

PHQ-9 and GAD-7 Scoring: Bands, Cutoffs and Change Over Time

The PHQ-9 and GAD-7 are the two short screens most often given together, one for depression and one for anxiety over the same two weeks. This sheet is the scoring half: both keys, the severity bands, how to read a high score on one against a low score on the other, and what the totals do not tell you. The forms with the questions on them stay on their own two printable pages.

PrivateFree PDFRuns in your browser
PHQ-9 and GAD-7 scoring: two people at a table compare two completed questionnaires laid side by side, one pointing a pen between the two score columns.

The two measures side by side

PHQ-9 (depression)

Nine items mapped onto the DSM symptom criteria for a major depressive episode, each scored 0 to 3 for the last two weeks. Total 0 to 27. Built as the depression module of the Patient Health Questionnaire, the self-report version of PRIME-MD, and validated as a severity measure by Kroenke, Spitzer and Williams in 2001.

GAD-7 (anxiety)

Seven items on worry, restlessness and anxious anticipation, each scored 0 to 3 for the last two weeks. Total 0 to 21. Developed by Spitzer, Kroenke, Williams and Lowe in 2006 as a severity measure for generalized anxiety disorder, and later found to screen reasonably for panic, social anxiety and post-traumatic stress as well.

What the two share

The same two-week window, the same four response options from Not at all to Nearly every day scored 0 to 3, and the same closing question about difficulty at work, at home and with other people, which is recorded but never added to the total. Both are free to use with no permission required, and both are severity measures rather than diagnostic tests.

Where the questions are

This sheet carries no questionnaire items. The two printable forms do: the PHQ-9 PDF and the GAD-7 PDF, each with the published wording, the response columns and the developers' attribution. Both sit under Tools on Psychology.com. Print those to fill in, and keep this one next to them to score.

How to score the PHQ-9

  1. Score each of the nine items: Not at all = 0, Several days = 1, More than half the days = 2, Nearly every day = 3.
  2. Add the nine item scores. The total runs from 0 to 27.
  3. Mark the band. The published cutpoints are 5, 10, 15 and 20: 0 to 4 none to minimal, 5 to 9 mild, 10 to 14 moderate, 15 to 19 moderately severe, 20 to 27 severe.
  4. Treat 10 or above as the commonly used positive-screen threshold that calls for further assessment, not as a diagnosis.
  5. Leave the closing difficulty question out of the total. It is not scored. Record it alongside the total, because it carries the person's own view of how much the symptoms interfere.
  6. Read the ninth item, which asks about thoughts of death or self-harm, on its own and at any total. An answer other than the lowest one is followed up directly in the same visit. If you are scoring your own form, tell a clinician about it now.
  7. If an item is left blank, record the total as incomplete and say so, rather than counting the blank as a zero. This missing-answer instruction is an editorial safeguard, not part of the published scoring.

How to score the GAD-7

  1. Score each of the seven items on the same scale: Not at all = 0, Several days = 1, More than half the days = 2, Nearly every day = 3.
  2. Add the seven item scores. The total runs from 0 to 21.
  3. Mark the band. The published cutpoints are 5, 10 and 15: 0 to 4 minimal, 5 to 9 mild, 10 to 14 moderate, 15 to 21 severe.
  4. Treat 10 or above as the recommended point for a fuller evaluation of an anxiety disorder.
  5. Leave the closing difficulty question out of the total here too, and record it next to the score.
  6. If an item is left blank, record the total as incomplete rather than counting the blank as a zero. This missing-answer instruction is an editorial safeguard, not part of the published scoring.

Reading the two together

Both elevated

The most common pattern, because depression and anxiety symptoms overlap heavily and several items are close cousins of each other. Two high totals are a reason to ask which came first, which is more disabling now, and which one treatment should aim at first. They are not two diagnoses.

GAD-7 high, PHQ-9 low

Points the next question toward worry, panic, social anxiety or trauma rather than toward mood. Because the GAD-7 screens anxiety broadly without separating the conditions, the follow-up is a conversation about what the anxiety attaches to and when it started, not a label read off the total.

PHQ-9 high, GAD-7 low

Worth checking what is carrying the total. Sleep, appetite, energy and concentration items can be pushed up by physical illness, pain, a sleep disorder or medication with no mood disorder present. Look at which items scored, not only at the sum, before treating the number as depression severity.

Change over time

This is the main use of both forms. Score on a schedule, write the date next to every total, and compare. A drop of 5 points or more on the PHQ-9 is the figure usually cited for a meaningful change in one person, estimated as two standard errors of measurement by Lowe and colleagues in 2004 in a sample of 434 older adults from the IMPACT trial. That estimate came from one sample in one trial of late-life depression, so treat 5 points as a rule of thumb rather than a standard.

What these scores are not

Both forms are screening and severity measures. Neither is a diagnostic test, and a total is not a condition. The cutoffs exist because somebody had to draw a line through a continuous distribution, and every line is a trade-off: move it down and more people with the condition are caught along with more people who do not have it, move it up and the reverse happens. Levis, Benedetti and Thombs examined the PHQ-9 threshold of 10 across 58 studies and 17,357 participants in 2019 and found sensitivity of 0.88 and specificity of 0.85 at that cutoff in the studies using semistructured interviews. Strong for nine questions, and still not a verdict on any individual person.

A depression diagnosis also means weighing things a questionnaire cannot see: grief, which can sit alongside depression, a history of manic or hypomanic episodes, and physical conditions, medications or other substances that produce the same symptoms. The GAD-7 has its own version of this limit, flagging anxiety broadly without distinguishing generalized anxiety from panic, social anxiety or post-traumatic stress. Use the totals the way a clinician does, as one input next to a history, a time course and what the person actually says.

Where the forms are

The two printable forms carry the questions, the response columns and the developers' attribution, ready to print and fill in by hand or take to an appointment. They are the PHQ-9 PDF and the GAD-7 PDF, both under Tools on Psychology.com.

If scoring by hand is not the point for you, the same two sets of questions are available as self-scoring versions, the Depression Test and the Anxiety Test, which produce the total and the band for you. Either way, keep this sheet for the interpretation, and keep the dates.

A record to keep

DateWrite the date beside every total. A single score says much less than two scores a month apart, and an undated total is almost useless later.
PHQ-9 total and bandRecord the 0 to 27 total, then the band it falls in, and note which items carried the total, because sleep, appetite, energy and concentration can be pushed up by things other than mood.
GAD-7 total and bandRecord the 0 to 21 total, then the band, and note what the worry attaches to, since the GAD-7 screens anxiety broadly without separating the conditions.
What the pair looks likeWrite one line on how the two totals sit against each other: both elevated, one clearly higher, or both low. That relationship is the part a clinician can use.
Change between datesNote the difference between the earlier total and the later one for each measure. On the PHQ-9, around 5 points is the rule of thumb for a change larger than ordinary measurement wobble, not a standard.
What to raise at the appointmentList the one thing the totals do not show: an item that has not moved, a side effect, a week that was unlike the rest of the month.

Source and permission

Both instruments were developed by Drs. Robert L. Spitzer, Janet B.W. Williams, Kurt Kroenke and colleagues, with an educational grant from Pfizer Inc. No permission is required to reproduce, translate, display or distribute them.

Primary citations: Kroenke K, Spitzer RL, Williams JBW. The PHQ-9: validity of a brief depression severity measure. Journal of General Internal Medicine. 2001;16(9):606-613. Spitzer RL, Kroenke K, Williams JBW, Lowe B. A brief measure for assessing generalized anxiety disorder: the GAD-7. Archives of Internal Medicine. 2006;166(10):1092-1097. Source of the PHQ-9 severity bands: Kroenke K, Spitzer RL. The PHQ-9: a new depression diagnostic and severity measure. Psychiatric Annals. 2002;32(9):509-515.

The scoring keys and severity bands on this page are reproduced under the developers' own statement in the official PHQ Screeners instruction manual that all of the PHQ family measures are in the public domain and that no permission is required to reproduce, translate, display or distribute them, with the developers' attribution retained. The questionnaire items themselves are not reproduced here. The PHQ family of measures and the official instruction manual are available free at https://www.phqscreeners.com.

Related tools & tests

Want to talk through what the score means?

Browse licensed therapists who treat depression and anxiety. Free to search, no account needed.

Find a therapist

Therapists Who Work With Depression

A worksheet is a start. These licensed therapists work with the same problem; every member is matched to the federal NPI Registry and board checks are dated on the profile.

  • Gladys LaiLCSW, MSW

    License verified, checked Oct 9, 2026

    New York, New York Telehealth $175 to $225 per session, sliding scale

    View profile
  • Gricelda AlvarezLMFT, MA

    License verified, checked Nov 9, 2026

    “Are you not sure if therapy is right for you? Let's talk and explore whether therapy might be beneficial to you. It is okay to ask for help.”

    San Diego, California Telehealth $120 per session

    View profile
  • Hanh DangLCSW

    License verified, checked Nov 10, 2026

    Long Beach, California Telehealth and in person $200 to $275 per session

    View profile

See all 199 depression therapists →

About this tool

PHQ-9 and GAD-7 scoring is usually the part people are actually hunting for. The two forms are everywhere, the questions take four minutes, and then someone has to turn the ticks into a number that means something. This page is that half of the job, and it deliberately carries no questionnaire items: the PHQ-9 form is at PHQ-9 PDF and the GAD-7 form is at GAD-7 PDF, each with the questions and response options as published. What follows is the scoring key for each, the severity bands, how to read the pair together, and the limits that go with both totals.

The two are given together because they were built to be. Both are distributed together as the Patient Health Questionnaire family, both ask about the last two weeks, both use the same four response options scored 0 to 3, and both finish with the same question about how much the problems have interfered with work, home and relationships, which is not added to the total. In primary care and in measurement-based mental health care, the pair is the standard short intake: a nine-item depression severity score out of 27 and a seven-item anxiety severity score out of 21, repeated at follow-up visits to show a direction of travel.

The PHQ-9 came out of PRIME-MD, a clinician-administered diagnostic procedure for primary care. Spitzer, Kroenke and Williams published the self-report version in JAMA in 1999, testing it on 3,000 adult patients across eight primary care clinics, with 585 of them interviewed by a mental health professional within 48 hours as the comparison standard. The nine items map onto the DSM symptom criteria for a major depressive episode. The validity paper for the nine-item depression module on its own is Kroenke, Spitzer and Williams in 2001, and the severity bands as they are usually printed are cited in full in the attribution at the end of this sheet.

The GAD-7 arrived later and from a different starting point. Spitzer, Kroenke, Williams and Lowe developed and validated it in 2006 as a brief severity measure for generalized anxiety disorder. A second study by Kroenke and colleagues in 2007, covering 965 patients across 15 primary care clinics, reported that it also performed reasonably as a screen for panic disorder, social anxiety disorder and post-traumatic stress disorder, with areas under the curve between 0.80 and 0.91. That is the useful and slightly awkward part of the GAD-7: a high score flags anxiety broadly rather than generalized anxiety specifically, and the form cannot tell those conditions apart.

If you are looking for a single sheet that covers both, the ultra-brief version already exists. The PHQ-4 combines the first two depression items with the first two anxiety items into a four-question screen, published by Kroenke, Spitzer, Williams and Lowe in 2009 on a sample of 2,149 primary care patients. It is scored 0 to 12 with a 3-or-above threshold on each two-item half. It is a triage tool rather than a severity measure: a positive PHQ-4 is a reason to give the full PHQ-9 or GAD-7, not a reason to record a severity level.

Who scores them, in practice, is whoever has the form: a medical assistant at check-in, a therapist at the start of a session, a care manager reviewing a caseload, or the person who filled it in. There is no training requirement and no scoring software needed, because the key is addition. The frequency is the part worth deciding in advance. Measurement-based care usually means scoring at intake and then at a fixed interval, every two to four weeks early in treatment and less often once someone is stable, with the date written next to every total. Scores collected on no schedule are hard to read later.

The ninth PHQ-9 item is handled differently from the other eight. It asks about thoughts of death or self-harm, and it is read on its own at any total, including a total in the minimal range. Any answer other than the lowest one is followed up directly, in the same visit, with questions about current thoughts, plans, means and supports, and a safety plan where one is needed. If you are scoring your own form, that follow-up is yours to start: tell a clinician about the answer now instead of waiting for the next score. That is also the single strongest argument against treating the PHQ-9 as a form someone fills in alone with no one reading it: the item exists so that an answer gets a response.

The PHQ-9 cutoff is a trade-off rather than a boundary in nature. The standard PHQ-9 threshold of 10 or above was examined by Levis, Benedetti and Thombs in a 2019 BMJ meta-analysis of individual participant data from 58 studies, 17,357 participants and 2,312 cases of major depression; in the 29 studies that used semistructured diagnostic interviews, sensitivity and specificity were best at that cutoff, at 0.88 and 0.85. Those are good numbers for a nine-question form and they still mean that a meaningful share of people are placed on the wrong side of the line. A cutoff sorts a population. It does not describe a person.

Which is the limit worth stating plainly: neither instrument was built for self-diagnosis, and a total is not a condition. A diagnosis of depression also means weighing grief, a history of manic or hypomanic episodes, and physical conditions, medications or other substances that could be producing the same symptoms, none of which a questionnaire can do. Grief and depression can also be present together. Low thyroid function, anemia, sleep apnea, chronic pain and several common medications all push PHQ-9 items upward on their own. The number is a well-made starting point for a conversation, and the interactive versions at Depression Test and Anxiety Test score themselves if that is easier than a printed form.

References

  1. Kroenke K, Spitzer RL, Williams JBW. The PHQ-9: validity of a brief depression severity measure. Journal of General Internal Medicine. 2001;16(9):606-613. doi:10.1046/j.1525-1497.2001.016009606.x
  2. Spitzer RL, Kroenke K, Williams JBW, Lowe B. A brief measure for assessing generalized anxiety disorder: the GAD-7. Archives of Internal Medicine. 2006;166(10):1092-1097. doi:10.1001/archinte.166.10.1092
  3. Spitzer RL, Kroenke K, Williams JBW. Validation and utility of a self-report version of PRIME-MD: the PHQ primary care study. JAMA. 1999;282(18):1737-1744. doi:10.1001/jama.282.18.1737
  4. Kroenke K, Spitzer RL, Williams JBW, Monahan PO, Lowe B. Anxiety disorders in primary care: prevalence, impairment, comorbidity, and detection. Annals of Internal Medicine. 2007;146(5):317-325. doi:10.7326/0003-4819-146-5-200703060-00004
  5. Kroenke K, Spitzer RL, Williams JBW, Lowe B. An ultra-brief screening scale for anxiety and depression: the PHQ-4. Psychosomatics. 2009;50(6):613-621. doi:10.1176/appi.psy.50.6.613
  6. Levis B, Benedetti A, Thombs BD. Accuracy of Patient Health Questionnaire-9 (PHQ-9) for screening to detect major depression: individual participant data meta-analysis. BMJ. 2019;365:l1476. doi:10.1136/bmj.l1476
  7. Lowe B, Unutzer J, Callahan CM, Perkins AJ, Kroenke K. Monitoring depression treatment outcomes with the Patient Health Questionnaire-9. Medical Care. 2004;42(12):1194-1201. doi:10.1097/00005650-200412000-00006

Cite this source

Psychology.com. (2026, December 4). PHQ-9 and GAD-7 Scoring. Psychology.com. https://psychology.com/tools/phq-9-and-gad-7

PHQ-9 and GAD-7 Scoring FAQ

How do you score the PHQ-9 and GAD-7?

PHQ-9 and GAD-7 scoring is addition. Each item is worth 0 for Not at all, 1 for Several days, 2 for More than half the days and 3 for Nearly every day, over the last two weeks. Add the nine PHQ-9 items for a total out of 27 and the seven GAD-7 items for a total out of 21. The closing question about difficulty at work, at home and with other people is recorded but never added to either total. Then mark the band: 5, 10, 15 and 20 are the PHQ-9 cutpoints, and 5, 10 and 15 are the GAD-7 cutpoints.

What do the PHQ-9 and GAD-7 cutoff scores mean?

They mark severity ranges, not diagnoses. On the PHQ-9, 0 to 4 is none to minimal, 5 to 9 mild, 10 to 14 moderate, 15 to 19 moderately severe and 20 to 27 severe. On the GAD-7, 0 to 4 is minimal, 5 to 9 mild, 10 to 14 moderate and 15 to 21 severe. A score of 10 or above on either one is the usual point at which a fuller assessment is recommended. Each cutoff is a trade-off between catching people who have the condition and flagging people who do not, which is why the same number means different things in different settings.

Can you use the PHQ-9 and GAD-7 for free?

Yes. Both were developed by Spitzer, Williams, Kroenke and colleagues with an educational grant from Pfizer Inc., and the developers state in the official PHQ Screeners instruction manual that the PHQ family of measures is in the public domain, with no permission required to reproduce, translate, display or distribute them and no fee of any kind. The condition in practice is that you keep the attribution with the form. That is why both printable versions on this site carry the developers' names and the primary citation at the foot of the page.

What does a high GAD-7 with a low PHQ-9 mean?

It points the next question toward anxiety rather than mood, and no further than that. The GAD-7 was built as a severity measure for generalized anxiety disorder, and Kroenke and colleagues reported in 2007 that it also screens reasonably well for panic disorder, social anxiety disorder and post-traumatic stress disorder, which means a high total flags anxiety broadly without saying which kind. The useful follow-up is what the anxiety attaches to, when it started and what it stops the person doing. A low PHQ-9 alongside it simply means the depression items were not endorsed for these two weeks.

How much of a change in PHQ-9 score is meaningful?

A drop of 5 points or more is the figure usually quoted for one person, estimated as two standard errors of measurement by Lowe and colleagues in 2004 in a sample of 434 older adults from the IMPACT depression trial. It is a rule of thumb rather than a standard, and it rests on one sample in one trial of late-life depression, which is a narrow base for a number now applied to everyone. The practical advice is unchanged. Score on a schedule, write the date next to every total, and read the direction rather than one number.

Can I diagnose myself with the PHQ-9 and GAD-7?

No, and neither form was built for it. Both are screening and severity measures, and a diagnosis also means weighing grief, a history of manic or hypomanic episodes, and physical conditions, medications or substances that produce the same symptoms. Low thyroid function, anemia, sleep apnea, chronic pain and several common medications can all push the PHQ-9 sleep, energy, appetite and concentration items upward on their own. What a score does well is organize what you want to say to a clinician: bring the total, the date, and which items you scored highest on.

Important: The PHQ-9 and the GAD-7 are screening and severity measures, not diagnoses, and this page is a scoring aid rather than medical advice. Any total, high or low, is a reason to talk with a clinician rather than a conclusion about what is happening.