EORTC QLQ-C30

EORTC QLQ-C30 in Oncology Trials and Economic Evaluation

The QLQ-C30 is the core questionnaire of the Quality of Life Group of the European Organisation for Research and Treatment of Cancer (EORTC), and cancer trials use it widely to measure health-related quality of life as patients report it. It reports a profile of separate functioning, symptom and global scores, each on a 0 to 100 scale, with no preference weights attached, so a health economist cannot use its scores as utilities for a QALY without a further step. That step is either a mapping function that predicts EQ-5D utilities from QLQ-C30 answers or the EORTC QLU-C10D, which scores a subset of the same items with national value sets. Aaronson and colleagues reported the instrument's international field study in 1993, in 305 patients with non-resectable lung cancer from centres in 13 countries, and found that completing the questionnaire took about 11 minutes on average.

Structure of the core questionnaire and its modules

The QLQ-C30 is a patient-reported outcome measure built as the core of a modular system. The EORTC Quality of Life Group describes the core as 30 items forming a number of scales, and disease-specific modules can be added to it for particular cancers or treatments.

Aaronson and colleagues described nine multi-item scales in the 1993 paper. Five are functional scales, three are symptom scales and one is a global health and quality of life scale. Six single items complete the questionnaire, so a full analysis produces 15 separate scores.

The response format differs between items. In version 3.0 the scoring manual codes most items from 1 (not at all) to 4 (very much), while the two global items, questions 29 and 30, are rated from 1 (very poor) to 7 (excellent).

Versions of the questionnaire and why they matter for pooled data

The scoring manual covers versions 1.0, (+3), 2.0 and 3.0 of the core questionnaire. The EORTC states that version 3.0 is the most recent and should be used for all new studies.

According to the scoring manual, version 3.0 differs from version 2.0 in giving four-point response scales to the first five items, which make up the physical functioning scale, and item 4 was reworded to fit the new categories. Earlier versions scored these items as yes or no. A physical functioning score from a version 1.0 trial is therefore built from a different response format, which matters when an evidence synthesis or a model pools QLQ-C30 data across older and newer trials.

Scoring each scale on a 0 to 100 range

Scoring starts with a raw score for each scale, the mean of the items that belong to it. The scoring manual allows a scale score when at least half of its items are answered, using the completed items, and then rescales every raw score linearly so that it runs from 0 to 100.

For functional scales a high score represents a high, healthy level of functioning, so the transformation reverses the item coding.

$$S_F = \left(1-\frac{RS-1}{r}\right) \times 100$$

where $S_F$ is the functional scale score, $RS$ is the raw score (the mean of the scale's item responses) and $r$ is the range of the item responses, which is 3 for items coded 1 to 4 and 6 for items coded 1 to 7.

For symptom scales, single items and the global scale the coding is kept in the same direction.

$$S_S = \frac{RS-1}{r} \times 100$$

where $S_S$ is the symptom, single-item or global score, and $RS$ and $r$ are defined as above. A high symptom score means a high level of symptoms or problems, while a high global score means better health and quality of life.

Illustrative worked example of two scale scores

The numbers below are illustrative and do not come from any trial. A patient answers the five physical functioning items with 1, 1, 2, 2 and 2, and rates the two global items 5 and 6.

1. Physical functioning raw score. The item total is 1 + 1 + 2 + 2 + 2 = 8, and the raw score is 8 / 5 = 1.6.

2. Physical functioning scale score. The range is 3, so (1.6 − 1) / 3 = 0.2, and (1 − 0.2) × 100 = 80.

3. Global scale raw score. The mean of the two global items is (5 + 6) / 2 = 5.5.

4. Global scale score. The range is 6, so (5.5 − 1) / 6 = 0.75, and 0.75 × 100 = 75.

A physical functioning score of 80 and a global score of 75 place this patient on two descriptive scales. Neither number is a utility of 0.80 or 0.75. On functional and global scales a score of 0 marks the worst response option and on symptom scales it marks no symptoms at all, so neither end of any scale stands for a state as bad as dead, and nobody's preferences between health states were used to set the distances between response levels.

Interpreting changes in QLQ-C30 scores

Differences in QLQ-C30 scores are judged against what patients regard as a meaningful change, which links the instrument to the idea of a minimal clinically important difference. Osoba and colleagues asked patients receiving chemotherapy to rate how much they had changed, alongside the QLQ-C30.

In that 1998 study, patients who reported a little change had mean score changes of about 5 to 10 points, those reporting a moderate change had changes of about 10 to 20 points, and those reporting very much change had changes of more than 20 points. The authors suggested these magnitudes could be used to calculate sample sizes for clinical trials. They describe groups of patients and do not turn a scale change into a QALY change.

The 15 separate scores also raise a multiplicity problem in trial reporting. Giesinger and colleagues tested higher-order models of the questionnaire in pretreatment data from 3,282 patients and reported in 2016 that a summary score from a single higher-order factor showed validity and responsiveness equal to, and in many cases better than, the original scale scores. The summary score is still a descriptive score and carries no preference weights.

Routes that turn QLQ-C30 data into utilities for cost-utility analysis

A cost-utility analysis needs a health state utility value for each state in a model, on a scale where full health is 1 and dead is 0. Rowen and colleagues noted that the QLQ-C30 is widely used in cancer but cannot be used in economic evaluation because it does not incorporate preferences.

Two routes have been developed. The first scores a selection of QLQ-C30 items through a preference-based classification. Rowen and colleagues derived an eight-dimension measure, often called the EORTC-8D, using psychometric analysis of trial data from 655 patients with multiple myeloma and a time trade-off valuation with 350 members of the UK general population. King and colleagues later developed the QLU-C10D, a ten-dimension classification that has since been valued with discrete choice experiments in many countries.

The second route is a mapping study that estimates a statistical relationship between QLQ-C30 answers and EQ-5D utilities in a dataset where patients completed both, and then applies it to a trial that collected only the QLQ-C30.

Mapping to the EQ-5D and the NICE methods

NICE's interim methods statement on the EQ-5D-5L value set (PMG51), published on 27 August 2026, says the UK EQ-5D-5L value set should be used to derive all utility values for adults in reference-case analyses, and it takes precedence over the NICE technology appraisal and highly specialised technologies manual (PMG36) until the manual is updated. Section 4.3.9 of PMG36 allows EQ-5D values to be estimated by mapping from other health-related quality of life measures when EQ-5D data are not available, and NICE is revising the relevant sections of the manual after PMG51. Section 1.10 of PMG51 says that data from a measure other than the EQ-5D should be mapped to utility values based on the UK EQ-5D-5L value set developed by Rowen and colleagues, for topics starting after the statement was published. Algorithms estimated before the EQ-5D-5L existed, such as those Versteegh and colleagues tested in 2010, predict EQ-5D-3L values and do not meet that requirement as they stand.

The NICE Decision Support Unit's Technical Support Document 22 replaces TSD 10 on mapping. The document says mapping needs an external dataset that records, for each patient and at the same time points, both the clinical measure used in the economic model and the EQ-5D. TSD 22 also cites a study by Lamu and Olsen in which linear regression produced predicted utilities above 1 when mapping from the QLQ-C30.

Mapping errors are not random across the range of health. Versteegh and colleagues reported in 2010 that algorithms mapping QLQ-C30 and arthritis-specific data onto the EQ-5D overpredicted utilities for patients in poor health. A separate algorithm for patients with an average QLQ-C30 functioning score below 45 performed better than existing algorithms but still did not predict mean utilities accurately. One implication for a cancer model, where progressed disease is often the poorest state, is that overprediction of that kind would narrow the utility gap between states.

How the QLQ-C30 differs from nearby measures

The QLQ-C30 is a disease-specific instrument, written for people with cancer, while the EQ-5D is a generic instrument that can be compared across conditions and has value sets that give utilities directly. The QLQ-C30 can detect symptoms such as nausea or constipation that a generic measure does not name, but its scores cannot be compared with outcomes in other disease areas on a common utility scale.

The QLU-C10D scores QLQ-C30 answers and has no questionnaire of its own, producing one utility from 13 of the 30 items according to the EORTC, although the QLU-C10D development paper by King and colleagues reported 12 items for its ten dimensions. Quality-adjusted time without symptoms (Q-TWiST) is an analysis of survival time split into periods with and without toxicity or progression, and it can draw its weights from many sources, while the QLQ-C30 supplies descriptive scores within those periods.

Limitations and common misreadings

Several errors recur when QLQ-C30 data reach an economic evaluation or a model. Most come from treating a descriptive profile as if it were a preference-weighted index.

• Reading a 0 to 100 score as a utility. A global score of 75 is not a utility of 0.75, because the scale is anchored on response options and not on full health and dead.
• Mixing score directions. A rise in a functional score is an improvement and a rise in a symptom score is a deterioration, so averaging or comparing raw changes across scales gives misleading results.
• Treating mapped values as observed. Mapped utilities carry prediction error, may exceed 1 under some models and overpredict in poor health, and that prediction error is part of the parameter uncertainty in the model.
• Pooling across versions. Physical functioning items changed response format between versions 2.0 and 3.0, so combined datasets need a check on which version each trial used.

Sources

• Aaronson NK, Ahmedzai S, Bergman B, et al. The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. Journal of the National Cancer Institute. 1993;85(5):365-376. doi:10.1093/jnci/85.5.365. https://pubmed.ncbi.nlm.nih.gov/8433390/
• EORTC Quality of Life Group. Core questionnaires (QLQ-C30). Accessed 17 October 2026. https://qol.eortc.org/questionnaires/core/
• Fayers PM, Aaronson NK, Bjordal K, Groenvold M, Curran D, Bottomley A, on behalf of the EORTC Quality of Life Group. The EORTC QLQ-C30 Scoring Manual. 3rd ed. Brussels: EORTC; 2001. https://www.eortc.org/app/uploads/sites/2/2018/02/SCmanual.pdf
• Giesinger JM, et al. Replication and validation of higher order models demonstrated that a summary score for the EORTC QLQ-C30 is robust. Journal of Clinical Epidemiology. 2016;69:79-88. https://pubmed.ncbi.nlm.nih.gov/26327487/
• King MT, Costa DSJ, Aaronson NK, et al. QLU-C10D: a health state classification system for a multi-attribute utility measure based on the EORTC QLQ-C30. Quality of Life Research. 2016;25(3):625-636. doi:10.1007/s11136-015-1217-y
• National Institute for Health and Care Excellence. Interim methods statement: implementing the EQ-5D-5L value set (PMG51). Published 27 August 2026. https://www.nice.org.uk/process/pmg51/chapter/statement-on-using-the-eq-5d-5l-value-set
• National Institute for Health and Care Excellence. NICE technology appraisal and highly specialised technologies guidance: the manual (PMG36). Published 31 January 2022, last updated 13 October 2026. https://www.nice.org.uk/process/pmg36/chapter/economic-evaluation-2
• Osoba D, Rodrigues G, Myles J, Zee B, Pater J. Interpreting the significance of changes in health-related quality-of-life scores. Journal of Clinical Oncology. 1998;16(1):139-144. doi:10.1200/JCO.1998.16.1.139
• Rowen D, Brazier J, Young T, et al. Deriving a preference-based measure for cancer using the EORTC QLQ-C30. Value in Health. 2011;14(5):721-731. doi:10.1016/j.jval.2011.01.004
• Versteegh MM, Rowen D, Brazier JE, Stolk EA. Mapping onto Eq-5 D for patients in poor health. Health and Quality of Life Outcomes. 2010;8:141. doi:10.1186/1477-7525-8-141
• Wailoo A, Hernandez Alava M, Pudney S. NICE DSU Technical Support Document 22. Mapping to estimate health state utilities. NICE Decision Support Unit; June 2023, updated March 2026 (version seen). https://sheffield.ac.uk/media/118671/download?attachment=