The EQ-5D, from the EuroQol group, describes perceived health across five dimensions. What it yields first is not a score but a five-digit code; the numeric index comes after, and it depends on whose country values you take. The code builder is right below.

The code depends on no country; the index does. The French value set has existed since 2020 and runs from -0.525 to 1, the negative part standing for states judged worse than death. The detail is below.

Figures taken from the studies listed at the end of this page, every value carrying its source where it is written. The questionnaire is not reproduced: its use goes through the EuroQol group.

What the EQ-5D measures

The EQ-5D describes perceived health across five dimensions: mobility, self-care, usual activities, pain or discomfort, and anxiety or depression. It adds a visual scale graduated from 0 to 100, where the patient places their health today.

It was published in 1990 by the EuroQol group, from postal surveys run in England, the Netherlands and Sweden. The result that founded the instrument is a striking similarity between the relative valuations those three populations attached to fourteen health states.

The original version scored each dimension on three levels. The five-level one, published in 2011, was built for an explicit reason: to gain sensitivity and reduce ceiling effects. That is the one to use today.

A health state code, then an index

This is the distinction almost every report skips, and it changes everything.

What the patient produces by answering is a five-digit code, one per dimension, from level 1 to level 5. The code 11223 describes a patient with no problem in mobility or self-care, slight trouble with usual activities, slight pain and moderate anxiety. That code depends on no country, no computation and no convention: it is the one to record in the notes.

The numeric index comes after. You get it by applying a value set to the code, that is a table giving a value to each of the 3,125 possible states, built by asking a country's population what they would give up to avoid this or that state. The index is therefore not a measure of the patient, but of the patient as seen through a population's preferences.

The index depends on the country

France has had its value set since 2020, built on 1,048 residents interviewed face to face. It runs from -0.525 to 1, the negative part standing for states the surveyed population judges worse than death.

That set does not rank the dimensions the way the three-level one did. Pain and mobility weigh most there, whereas the earlier version put mobility and self-care first. The same patient, described by the same code, therefore does not get the same index depending on the version and on the country of the value set.

The practical consequence fits in one line: an EQ-5D index written without saying which value set it comes from cannot be interpreted, and two studies using different sets do not compare. The code, by contrast, always compares.

The threshold for change is not a constant

There is no single minimally important difference for the EQ-5D, and claiming one would be a mistake.

The 47-article review published in 2024 records thresholds running from 0.01 to 0.41 for the five-level index, and from 0.42 to 23 points for the visual scale. Two factors explain most of that spread: surgical intervention and a low baseline score both push the threshold up. Its authors conclude that a uniform threshold is not appropriate.

Studies by population confirm it. In 758 patients with hip or knee osteoarthritis, the responsiveness of the instrument is small among non-surgical patients, while the improvement threshold reaches 0.32 points among surgical patients who get better. In other words, the EQ-5D sees surgery well and rehabilitation poorly.

Three administration pitfalls

Writing an index without saying where its values come from

An index of 0.72 means nothing on its own. The report must carry the health state code and, if the index is computed, name the value set and the version used.

Still using the three-level version

In 146 patients with low back pain who filled in both at the same time, the five-level version describes 73 health states against 28, with markedly lower ceiling effects. Its authors recommend preferring the five-level version.

Expecting the EQ-5D to follow a course of rehabilitation

It is a generic instrument, designed for economic evaluation, not for measuring functional progress from session to session. A region-specific scale will do that better, and the EQ-5D will stay useful alongside, for what it can do.

Frequently asked questions

What does EQ-5D stand for?

EQ-5D is the name of the instrument from the EuroQol group, published in 1990. The 5D refers to its five dimensions: mobility, self-care, usual activities, pain or discomfort, anxiety or depression.

How is the EQ-5D scored?

Each dimension gets a level, from 1 to 5 in the current version, which forms a five-digit code. A numeric index can then be derived from it, but that requires a country's value set.

How much change counts?

There is no single answer. Published thresholds run from 0.01 to 0.41 for the five-level index depending on treatment type and baseline score, and a uniform threshold is not appropriate.

Is there a French value set?

Yes, published in 2020 from 1,048 French residents. It covers the 3,125 possible states and runs from -0.525 to 1.

Is the EQ-5D free to use?

No. Its use goes through registration with the EuroQol group, which holds the rights and issues the translated versions. That is why this page does not reproduce its questionnaire.

References

6 sources, PMIDs included

1. EuroQol Group. EuroQol: a new facility for the measurement of health-related quality of life. Health Policy 1990;16(3):199-208. PMID 10109801. The original publication, by postal surveys in England, the Netherlands and Sweden. It records a striking similarity between the relative valuations attached to 14 health states by different populations, measured on a visual analogue scale.
2. Herdman M, Gudex C, Lloyd A, Janssen M, Kind P, Parkin D, Bonsel G, Badia X. Development and preliminary testing of the new five-level version of EQ-5D (EQ-5D-5L). Qual Life Res 2011;20(10):1727-36. PMID 21479777. The five-level version, created for an explicit reason: to gain sensitivity and reduce the ceiling effects of the three-level version. The intermediate levels are slight, moderate and severe, between the anchors 'no problems' and 'unable to do'.
3. Andrade LF, Ludwig K, Goni JMR, Oppe M, de Pouvourville G. A French Value Set for the EQ-5D-5L. Pharmacoeconomics 2020;38(4):413-25. PMID 31912325. The French value set, built on 1,048 residents interviewed face to face. It gives a value to each of the 3,125 possible states, from -0.525 to 1. The ranking of dimensions changes there: pain and mobility weigh most, whereas the three-level version put mobility and self-care first.
4. Cheng LJ, Chen LA, Cheng JY, Herdman M, Luo N. Systematic review reveals that EQ-5D minimally important differences vary with treatment type and may decrease with increasing baseline score. J Clin Epidemiol 2024;174:111487. PMID 39084578. The 47-article review that forbids quoting a single threshold. Minimally important differences run from 0.01 to 0.41 for the five-level index and from 0.42 to 23 points for the visual scale. The authors conclude that a uniform threshold is not appropriate.
5. Bilbao A, García-Pérez L, Arenaza JC, García I, Ariza-Cardiel G, Trujillo-Martín E, Forjaz MJ, Martín-Fernández J. Psychometric properties of the EQ-5D-5L in patients with hip or knee osteoarthritis: reliability, validity and responsiveness. Qual Life Res 2018;27(11):2897-908. PMID 29978346. 758 patients with hip or knee osteoarthritis, seen again at six months. Floor and ceiling effects below 3 %, internal consistency 0.86, strong correlations with the WOMAC. Responsiveness is small among non-surgical patients and the improvement threshold settles at 0.32 points among improved surgical patients.
6. Garratt AM, Furunes H, Hellum C, Solberg T, Brox JI, Storheim K, Johnsen LG. Evaluation of the EQ-5D-3L and 5L versions in low back pain patients. Health Qual Life Outcomes 2021;19(1):155. PMID 34049574. 146 patients with low back pain who filled in both versions at the same time. The five-level version describes 73 health states against 28 for the three-level one, and its ceiling effects are markedly lower on mobility, self-care, pain and anxiety. The authors recommend it.