To score the WHO-5 Wellbeing Index, add the five item ratings (each 0 to 5) for a raw score of 0 to 25, then multiply by 4 for a percentage score of 0 to 100. Higher means better wellbeing. A percentage below 50 (raw below 13) suggests poor wellbeing and warrants further assessment, for example for depression.
That is the official rule, and it comes from the World Health Organization's 2024 edition of the WHO-5. What most scoring pages miss is that the 2024 edition quietly dropped two rules printed on the older form many practices still use. One of them is the "10% change" rule that clinicians often rely on to call improvement.
How do you score the WHO-5?
The WHO-5 asks how the person has felt over the last two weeks. Each of the five statements is rated on a six-point scale: 5 for "All of the time", 4 for "Most of the time", 3 for "More than half of the time", 2 for "Less than half of the time", 1 for "Some of the time" and 0 for "At no time".
The five statements cover feeling cheerful and in good spirits, calm and relaxed, active and vigorous, waking up fresh and rested, and having a daily life filled with things of interest. All five are positively worded, so nothing is reverse-scored.
- Add the five ratings. The raw score runs from 0 (worst possible wellbeing) to 25 (best possible).
- Multiply the raw score by 4. The percentage score runs from 0 to 100.
Worked example: a client rates the items 3, 2, 2, 1 and 3. The raw score is 11 and the percentage score is 44. Because 11 is below 13, this result meets the WHO's threshold for further assessment.
If an item is missing, there is no official pro-rating rule in the WHO document, so record the score as incomplete rather than guessing.
What does a WHO-5 score mean?
The 2024 WHO edition gives one interpretive threshold: a percentage score below 50, or a raw score below 13, "has been suggested as a cut-off for poor mental well-being" and as a prompt for further assessment for a possible mental health condition such as depressive disorder.
That is the whole official interpretation. Unlike the PHQ-9 or the K10, the WHO-5 has no published mild, moderate and severe bands from its developers. If you see a page that splits WHO-5 scores into four or five severity levels, those bands come from somewhere other than the WHO.
Keep the direction in mind too. The WHO-5 measures positive wellbeing, so a low score is the concerning one. That is the reverse of most symptom scales, and it is an easy slip when one client completes the WHO-5, PHQ-9 and GAD-7 in the same session.
What is the WHO-5 cut-off for depression?
Two cut-offs circulate: raw below 13 (percentage below 50) and raw 7 or less (percentage 28 or less). The second is the origin of the "28%" figure that appears on many pages.
The best Australian evidence on the choice comes from Halliday and colleagues (2017), who tested both cut-offs against the PHQ-9 in 3,249 Australian adults with diabetes from the Diabetes MILES study:
- Raw below 13: sensitivity 0.79, specificity 0.79.
- Raw 7 or less: sensitivity 0.44, specificity 0.96.
In plain terms, the stricter cut-off of 7 missed more than half of the people with likely depression. The authors supported the below-13 cut-off. Note the population: adults with diabetes, with the PHQ-9 rather than a diagnostic interview as the comparison. It is a strong Australian data point, not a universal answer.
The broader literature points the same way. A systematic review of 213 studies by Topp, Østergaard, Søndergaard and Bech concluded that the WHO-5 is a sensitive and specific screening tool for depression. A positive screen is still a reason to assess, not a diagnosis.
What changed in the 2024 WHO edition?
In 2024 the Psychiatric Centre North Zealand in Denmark, where the late Per Bech developed the WHO-5, assigned its copyright to the WHO, which republished it as an open access product.
The archived 1998-version form from the Danish centre carried three interpretation rules:
- A raw score below 13 indicates poor wellbeing and is an indication for testing for depression.
- Depression assessment is also recommended if the person answered 0 or 1 to any of the five items, whatever the total.
- To monitor change, use the percentage score, where "a 10% difference indicates a significant change".
The 2024 WHO edition keeps the first rule. It does not repeat the second or the third.
The item-level rule matters clinically. A client can score 15 overall while answering "At no time" to waking up fresh and rested. On the old form that triggered further assessment. On the 2024 form, the total alone would not. Nothing stops you from following up a 0 or 1 on any item, and many clinicians will, but it is now clinical judgement rather than a printed scoring rule.
How much does a WHO-5 score need to change?
The older form's 10-point rule is widely repeated, but test-retest studies suggest that a 10-point change can sit inside ordinary measurement noise.
Two Danish studies estimated the minimal detectable change, the smallest shift that exceeds measurement error for an individual:
- Schougaard and colleagues (2018), in 554 adult epilepsy outpatients (with Bech as a co-author), found a minimal detectable change of 23.6 points.
- Schougaard and colleagues (2022), in 146 adults with type 1 diabetes completing it online, found 18.6 points and described the WHO-5's measurement error as large.
Both samples were medical rather than mental health populations, so treat these as cautionary figures, not settled thresholds. A practical reading:
- A change of 10 points or less: do not call it improvement or deterioration on the WHO-5 alone.
- A change of roughly 20 points or more: likely real change for that individual.
- In between: look for corroboration from a symptom measure, the client's own account and functioning.
Using the worked example above, a client moving from 44 to 56 has improved by 12 points. That met the old 10% rule but sits below both minimal detectable change estimates. Record it as possible improvement and check it against another measure.
There is Australian support for the WHO-5 as a monitoring tool. Newnham, Hooke and Page (2010) used it routinely with 318 patients at an Australian private psychiatric facility and found it reliable, valid and sensitive to change in people with affective and anxiety disorders, with early scores predicting final outcome.
Can you use the WHO-5 for a Mental Health Treatment Plan?
Yes, if it is the right tool for the client. The MBS descriptor for item 2715 requires the GP to administer an outcome measurement tool when preparing a Mental Health Treatment Plan (except where clinically inappropriate), leaves the choice of evidence-based tool to clinical discretion, and names the K10 and the DASS-21 as examples. The WHO-5 is not named, but it is not excluded.
The catch is the review. The same descriptor requires "re-administration of the outcome measurement tool used in the assessment stage". If the plan was built on a K10, a WHO-5 at review does not satisfy that requirement. Our guide to Mental Health Treatment Plan referrals covers the rest of the process.
Many psychologists pair the WHO-5 with a symptom measure for this reason. The symptom measure answers "is this getting better?" and the WHO-5 answers "is life getting better?", which are not always the same question.
Is the WHO-5 free to use?
The 2024 WHO edition is published under a Creative Commons CC BY-NC-SA 3.0 IGO licence, which allows copying and adaptation for non-commercial purposes with attribution. The WHO publication page lists more than 25 translations, and the WHO directs requests for commercial use to its copyright page. If you plan to build the WHO-5 into a commercial product, check the licence terms rather than assuming.
How should you record WHO-5 results in clinical notes?
A WHO-5 entry that will still make sense at review usually includes:
- the raw score and the percentage score, labelled, because "13" and "52" can describe the same result;
- any individual items scored 0 or 1;
- whether the result is below the 50% threshold, and what follow-up was done;
- the change since the last administration, and whether another measure supports it.
Scoring by hand is where errors creep in, especially the multiply-by-four step and mixing up raw and percentage scores across sessions. PractaLuma is AI-native practice management software for Australian mental-health practices, and its standardised assessments include the WHO-5 alongside the PHQ-9, K10 and DASS-21, so results are kept in the client record rather than on loose paper forms. See pricing for plan details.
Frequently asked questions
What is a normal WHO-5 score?
There is no official "normal" range. The WHO's only threshold is a percentage below 50 (raw below 13), which suggests poor wellbeing. Population norms exist for some countries, such as a 2025 German representative sample Interpret any norm against the population it came from.
What does a WHO-5 score of 7 mean?
A raw score of 7 is a percentage score of 28. It is well below the WHO's threshold of 13 and at the stricter cut-off some studies use for likely depression. It warrants a proper assessment, including risk.
Is the WHO-5 a depression test?
No. It measures positive mental wellbeing over the last two weeks. It works reasonably well as a first-step screen for depression, but a low score needs a clinical assessment before any diagnosis.
Should I use the raw score or the percentage score?
Either, as long as you label it. The WHO cut-off is stated both ways (raw below 13, percentage below 50). Most change research reports the percentage scale, so use it for tracking over time.
This article is general information for clinicians and is not a substitute for clinical judgement.