PHQ-9 Depression Severity Calculator
PHQ-9 Depression Severity Calculator
Score the nine PHQ-9 items out of 27 against Kroenke’s published severity bands. The ninth item asks about self-harm and is shown separately, because it is not a risk assessment and a total is not either.
PHQ-9 total
9 items, 0 to 27Anhedonia 2; depressed mood 2; sleep 1; energy 2; appetite 1; worthlessness 1; concentration 1; retardation or agitation 1; item 9 not at all (0)
Scoring
Range 0 to 27 · bands at 5, 10, 15 and 20 · screening cut-off 10 or more
- the ninth item is not a risk assessment
- it asks how often a thought of being better off dead or of self-harm has been present over two weeks, and nothing about intent, plan, means, access or history. Any answer above not at all needs direct clinical assessment of the person rather than a score, and the answer is shown beside the total above for exactly that reason
- a total cannot show an absence of risk
- the nine items are summed with equal weight, so someone answering nearly every day to the ninth item and not at all to the other eight scores 3 and lands in the lowest band. No total on this page can be read as showing low risk of self-harm, as excluding it, or as showing that assessment is unnecessary
- 88 and 88, at a cut-off of 10
- Kroenke, Spitzer and Williams 2001: 6,000 patients in 8 primary care and 7 obstetrics-gynaecology clinics, with an independent structured interview by a mental health professional in 580 as the criterion standard. A total of 10 or more gave 88 per cent sensitivity and 88 per cent specificity for major depression
- what the bands are actually for
- severity tracking and treatment-response monitoring, which is what a repeated ordinal total does well. They are a convention over round numbers, not diagnostic categories, and a score is not a diagnosis of a disorder
- two items are bidirectional
- items 3 and 5 score sleep and appetite the same whether they are reduced or increased, and item 8 scores retardation and agitation together. Three of the nine points can therefore be earned by opposite clinical states, and the total cannot show which
Worked example
Anhedonia 2; depressed mood 2; sleep 1; energy 2; appetite 1; worthlessness 1; concentration 1; retardation or agitation 1; item 9 not at all (0)
2 + 2 + 1 + 2 + 1 + 1 + 1 + 1 + 0 = 11 points
11 sits in Kroenke's 10-to-14 band and at or above the screening cut-off of 10
Apply the derivation cohort's 88 per cent sensitivity and 88 per cent specificity to 1,000 people at a major-depression prevalence of 7 per cent: 70 with the disorder give 62 true positives, and 930 without give 112 false positives. 62 of 174 is a positive predictive value of 35.6 per cent
At a prevalence of 20 per cent the same two figures give 64.7 per cent. The instrument is unchanged; the population is not
Now the arithmetic that matters more. Set every item to not at all and the ninth item to nearly every day: the total is 3, which falls in the lowest band on the bar above. Read the ninth item on its own: it is not a risk assessment, it needs direct clinical assessment of the person rather than a score, and no total here can be read as showing low risk of self-harm.
The published severity bands, and what is behind them
| Total | Kroenke 2001 label | What is attached to it |
|---|---|---|
| 0 to 4 | Below the mild threshold | Nothing beyond the convention |
| 5 to 9 | Mild | Nothing beyond the convention |
| 10 to 14 | Moderate | The cut-off of 10: sensitivity 88 per cent, specificity 88 per cent for major depression |
| 15 to 19 | Moderately severe | Nothing beyond the convention |
| 20 to 27 | Severe | Nothing beyond the convention |
What a PHQ-9 of 10 or more is worth, by prevalence
| Prevalence of major depression | Positive predictive value | Negative predictive value |
|---|---|---|
| 20 in 100 | 64.7 per cent | 96.7 per cent |
| 7 in 100 | 35.6 per cent | 99.0 per cent |
| Sensitivity used | 88 per cent (Kroenke 2001) | |
| Specificity used | 88 per cent (Kroenke 2001) |
A severity scale with one item that must be read on its own
The PHQ-9 is the depression module of the Primary Care Evaluation of Mental Disorders, scored as nine items of 0 to 3 over the last two weeks. Kroenke, Spitzer and Williams validated it in 2001 in 6,000 patients across 8 primary care clinics and 7 obstetrics-gynaecology clinics, against an independent structured interview by a mental health professional in a subsample of 580. A total of 10 or more gave 88 per cent sensitivity and 88 per cent specificity for major depression, and the paper proposed scores of 5, 10, 15 and 20 as the thresholds for mild, moderate, moderately severe and severe depression.
Those four thresholds are a convention and not a diagnosis. Only one of them — the cut-off of 10 — has an accuracy pair attached, and even that pair weakens sharply once it meets a real population: at the 7 per cent major-depression prevalence of primary care, a total of 10 or more carries a positive predictive value of about 36 per cent. What the ordinal total is good at is the thing it is least often used for, which is following one person across repeated administrations: a fall from 18 to 9 in the same patient means something a single 9 does not.
Three structural properties change how a total should be read. Items 3 and 5 score sleep and appetite identically whether they are increased or decreased, and item 8 scores retardation and agitation together, so three of the nine points can be earned by opposite clinical states. Items 3, 4 and 5 are the somatic ones, and in a medically unwell population they can produce a total of 6 with no mood symptom at all. And the ninth item is categorically unlike the other eight.
The ninth item. It asks how often thoughts of being better off dead or of self-harm have been present over two weeks. The ninth item is not a risk assessment. It asks how often a thought has been present over two weeks and nothing about intent, plan, means or history, and any answer above not at all needs direct clinical assessment of the person rather than a score. No total on this page can be read as showing low risk of self-harm, as excluding it, or as showing that assessment is unnecessary. That inference is the dangerous one: the nine items are summed with equal weight, so someone who answers nearly every day to the ninth item and nothing at all to the other eight scores 3 and sits in the lowest band on the bar above. That is why this page shows the ninth item’s own answer beside the total and says so again whenever it is above zero. If you are reading this about yourself and you are in distress or thinking about harming yourself, please contact your local emergency number or a crisis line now rather than treating a number as an answer: 999 or Samaritans on 116 123 in the UK and Ireland, 988 in the United States and Canada, or your local emergency service elsewhere.
Attribution, as the instrument’s own notice requires. Developed by Drs Robert L. Spitzer, Janet B.W. Williams, Kurt Kroenke and colleagues, with an educational grant from Pfizer Inc. The published form carries “No permission required to reproduce, translate, display or distribute”, and that notice is the reason this page exists while no page here reproduces the AUDIT.
A low total does not exclude what the instrument screens for. No sensitivity quoted here is 1, every figure was measured in a published cohort rather than in the person in front of you, and someone who endorses nothing may still have the disorder. This page renders no dose, no medication, no treatment regimen and no disposition. It computes the published total and prints the published thresholds with the body that published each one; what follows from the number is a clinical decision this page does not make. Every cut-off, sensitivity, specificity and severity band here is printed with the cohort it was measured in, or with a statement that no source read for this page attaches one — because an instrument’s accuracy is a property of the population it was measured in and not of the instrument.
Frequently asked questions
What do the PHQ-9 severity bands mean?
Kroenke’s 2001 paper put scores of 5, 10, 15 and 20 forward as the thresholds for mild, moderate, moderately severe and severe depression. They are a severity convention over round numbers. Only the cut-off of 10 has diagnostic accuracy attached to it — 88 per cent sensitivity and 88 per cent specificity in the derivation cohort — and a total on this page is a number, not a diagnosis. These instruments quantify what a person reports, or what an observer records at one moment; a diagnosis rests on a clinical assessment that no questionnaire total stands in for.
What does a positive answer to PHQ-9 item 9 mean?
The ninth item is not a risk assessment. It asks how often a thought has been present over two weeks and nothing about intent, plan, means or history, and any answer above not at all needs direct clinical assessment of the person rather than a score. If you are reading this about yourself and you are in distress or thinking about harming yourself, please contact your local emergency number or a crisis line now rather than treating a number as an answer: 999 or Samaritans on 116 123 in the UK and Ireland, 988 in the United States and Canada, or your local emergency service elsewhere.
What does a low PHQ-9 total say about the risk of self-harm?
No total on this page can be read as showing low risk of self-harm, as excluding it, or as showing that assessment is unnecessary. That inference is the dangerous one: the nine items are summed with equal weight, so someone who answers nearly every day to the ninth item and nothing at all to the other eight scores 3 and sits in the lowest band on the bar above.
How accurate is a PHQ-9 of 10 or more?
In the derivation cohort — 6,000 primary care and obstetrics-gynaecology patients, with 580 given an independent structured interview by a mental health professional — it carried 88 per cent sensitivity and 88 per cent specificity for major depression. At a prevalence of 7 per cent that works out to a positive predictive value of 35.6 per cent, and at 20 per cent to 64.7 per cent.
Is the PHQ-2 enough on its own?
The PHQ-2 is items 1 and 2 of this instrument, scored 0 to 6, with a cut-off of 3 or more. In the 2003 validation it carried 82.9 per cent sensitivity and 90.0 per cent specificity for major depressive disorder. It is designed as a first stage, and a reader following that route should note that the ninth item is not in it.
Related calculators
References
- Patient Health Questionnaire (PHQ-9) and Generalized Anxiety Disorder 7-item scale (GAD-7), the combined published form as distributed by the North Dakota Department of Health and Human Services. Source of every item of both instruments in its exact wording, the four response options and their point values, the shared functional-impairment question, and the developer notice: “Developed by Drs. Robert L. Spitzer, Janet B.W. Williams, Kurt Kroenke and colleagues, with an educational grant from Pfizer Inc. No permission required to reproduce, translate, display or distribute, 1999.”
- Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606–613. Abstract read: 6,000 patients in 8 primary care clinics and 7 obstetrics-gynaecology clinics; the criterion standard an independent structured interview by a mental health professional in a subsample of 580; “PHQ-9 scores of 5, 10, 15, and 20 represented mild, moderate, moderately severe, and severe depression, respectively”; and a score of 10 or above carrying 88 per cent sensitivity and 88 per cent specificity for major depression.
- Patient Health Questionnaire-9 (PHQ-9), National HIV Curriculum, University of Washington. Second independent reproduction of all nine items and the four response values, the 0-to-27 range, the five severity bands at 0-4, 5-9, 10-14, 15-19 and 20-27, and the statement that a patient who answers yes to question 9 needs further assessment for suicide risk by a qualified clinician.
Not medical advice. For healthcare professionals and education. Reference intervals vary by laboratory and assay — always use your own laboratory's. Never base a dose or a treatment decision on this page alone. Full disclaimer at calcengines.com/disclaimer/