First of all, thank you for your interest in the EORTC QLQ-C30 and/or its modules and for contacting us.

Generally the Quality of Life Department does not involve itself in external research projects. We receive numerous requests, on a daily basis and it is impossible for us to design a trial, provide actual data-management or to perform analyses for free as a consultancy service. Neither do we take any responsibility for results obtained outside of EORTC framework. Hopefully, the information provided on the FAQ web page will help you consider some of the basic issues in designing clinical trials and research projects. In some instances, we can offer consultancy services for a fee and requests for this should be initially directed to the Head of the Quality of Life Department, Dr Andrew Bottomley.

Should you be interested in starting a new research project under the auspices of the EORTC Quality of Life Group/ or Department, please contact the chairman of the group or the Head of the Quality of Life Department.

There are 4 main steps to consider for the design of a QoL study:

• Determine your QOL Objective
  • Write down the specific question you want to answer with your question. Take extra note to determine which specific QoL domains, what time periods and what type of changes/differences you are expecting to be relevant for your question.
• Choose an instrument
  • It should be a validated (ie. reliable, valid, responsive, feasible ?) questionnaire in the trial population.
  • It should address the QoL domains that are essential for your hypothesis. global measures, disease-specific measures, symptom checklists
  • Check if appropriate translations are available.
• Select your assessment time points.
  • Assessment schedule should closely reflect the objective.
  • Should include at least baseline (ie. before treatment/randomization).
  • If comparing multiple arms: schedule should be identical/comparable between the arms.
  • Keep the patient in mind: do not overburden by collecting more than necessary.
• Develop an analysis plan
  • Analysis plan should reflect the objective.
  • Involve address issues of scale/domain selection, sample size, compliance, missing data, minimal clinically important difference and sensitivity analyses.

Note that objectives, instrument, assessment schedule and analysis techniques should be specified a priori. A good trial design, conduct, analysis and reporting is only possible if these aspects have been determined beforehand.

In order to obtain the QLQ-C30 and modules you need to ask for permission. This can be done via our download section. By completing the process you are authorised to use the questionnaire in that specific study. For each new study, you need to complete another request. There is no fee for academic or non-commercial users but if your study has commercial sponsorship you will be required to pay a fee dependent upon the number of patients in the study. By clicking here you will be able to fill in a request. You will receive an automated reply with the links where you can download the documents you have requested. Clicking here will enable you to fill in a user's agreement.

Modules that have been validated, are cited on this page. For unvalidated modules please contact Ken Cornelissen.

Details of available translations can be found here or by contacting the Translation coordinator.

More information on getting translations done can be obtained from the Translation coordinator. The Translation manual can also be of help.

At present we are developing a module for chronic lymphocytic leukaemia (CLL). In the past, other modules may have been developed for Leukaemia and indeed other disease sites, but these are not in the formal repertoire of EORTC modules unless they are listed on our web site. If you are unsure if the module you need is a formal module, please see our modules section for a list of our present modules.

In order to be able to use the QLQ-C30 and modules you need to ask for permission. This can be done via our download section. By completing the process you are authorised to use the questionnaire in that specific study. For each new study, you need to complete another request. There is no fee for academic or non-commercial users but if your study has commercial sponsorship you will be required to pay a fee dependent upon the number of patients in the study. By clicking here you will be able to fill in a request. You will receive an automated reply with the links where you can download the documents you have requested.

You do, but this is a simple process as described above.

No, only if your study has commercial sponsorship you will be required to pay a fee, dependent upon the number of subjects in the study. As an academic user, you will however also need to register to be able to use our questionnaires.

On the download page you can select the questionnaire you need (eg. QLQ-C30, QLQ-C15-PAL or IN-PATSAT32). A new screen opens where you should fill in your details (You must fill in each area, if you do not have a fax number, please put your tel. number into that area), once you have completed them at the bottom of this page click on SUBMIT. On the next page please select the documents you require (questionnaires, languages, full reference values manual, or parts of the manual). At the bottom of the page you need to tick the box that you agree to the terms of the User's agreement. By clicking the link the user's agreement opens in a new screen. If you have done that, please click on SUBMIT and then your request will be sent through. You will automatically receive an email with the download details of the documents you requested. Therefore it is essential that you have filled in your proper email address. So please check it carefully to avoid any inconvenience.

For more information on this subject, please review the validation pages.

Translations are conducted using a monitored forward/backward procedure to ensure good quality. More information on this procedure can be found in the translation guidelines.

All items in the module should be used and the order of the items can not be changed. The questionnaire has been developed and validated in this format and changing the order of the items might affect the psychometric properties of the questionnaire.

All items in the module should be used and the order of the items can not be changed. The questionnaire has been developed and validated in this format and changing the order of the items might affect the psychometric properties of the questionnaire.

You will be sent a scoring manual when you complete a user's agreement for the QLQ-C30. In addition to the algorithms and worked examples, the manual contains syntax for common computer software packages (SPSS, SAS). The syntax has also been reproduced on a CD ROM which can be obtained from the EORTC Quality of Life Department by contacting Ken Cornelissen.

We do not recommend that, since we do not know the relative weights to put on each items/scale.

A general recommendation is provided in the scoring manual. Users should be careful to impute missing values if there are a lot of them-and especially, if they seem to be non-random. There might be a clinical reason/explanation of why patients have dropped a specific item. Volume 17, issue 5-7 (1998) of Statistics in Medicine contains a series of published papers on missing data in quality of life research in cancer clinical trials. References of interest are: Incomplete quality of life data in randomized trials: missing forms. Curran D, Molenberghs G, Fayers PM, Machin D. Stat Med. 1998 Mar 15-Apr 15;17(5-7):697-709 and Why are missing quality of life data a problem in clinical trials of cancer therapy? Fairclough DL, Peterson HF, Chang V. Stat Med. 1998 Mar 15-Apr 15;17(5-7):667-77.

This depends on whether the data is missing at random or systematically. The importance of keeping missing data to a minimum cannot be over-emphasized. A manual on 'Guidelines for assessment of Quality of Life in EORTC Clinical Trials' is available from the EORTC Quality of Life Department, which discusses ways of reducing missing data at all stages of a study, from design to implementation in the clinic. General recommendations concerning missing data can be found in the scoring manual.

In the EORTC, the following information concerning the different reasons for missing data is collected:

• patient felt too ill;
• clinician or nurse felt the patient was too ill;
• patient felt it was inconvenient, takes too much time;
• patient felt it was a violation of privacy;
• patient didn't understand the actual language / illiterate;
• administrative failure to distribute the questionnaire;
• not required at this time point;
• other, specify...;
• unknown.

When a response is missing, it should be coded as "9" for missing data (see scoring manual). When two adjacent categories have been circled by the patient, the category which represents the worst quality of life should be taken. When two categories which are not adjacent have been circled, then the response is not evaluable and it should be coded as "8".

The only scales that differ in the current (v3.0) and previous versions are the global health status/QoL (QL) scale, the physical functioning (PF) scale and the role functioning (RF) scale. The different scoring for these scales in the different versions has been summarized in the scoring manual.

For validated modules, the scoring procedures have been incorporated in the scoring manual. For the scoring of modules that have not yet been validated, the procedure depends on the development phase of the module. For those at phase III, the scoring will be supplied from the EORTC Quality of Life Department along with the user's agreement. For modules at phase I and II, the module developer should be consulted for the scoring. It is not allowed to publish psychometric data from any unvalidated modules (phase I, II or III) until the results of the module validation study have been published.

All the information needed to perform accurate scoring is provided in the scoring manual, which can be obtained from the EORTC Headquarters. The Quality of Life Department at the EORTC Headquarters can give additional advice on the scoring if needed.

Generally: no.
The current scoring system has been extensively tested and validated. By scoring the items differently, the resulting scales might no longer measure the concepts they were intended to measure.
You are allowed to perform linear transformations on the scales should that be helpful. E.g. you might rescale the scales from 0-100 to a 0-1 scale or change the direction of the scales if it would make the presentation of the results more comprehensible. If you do so, please mention clearly the transformation you used. Otherwise people familiar with the classical scoring might be misled.

Item 13 is indeed not included in a scale. The question refers to the use of concomittant medication. Such medication is not always allowed/available. The use of this item is therefore trial specific. The analysis and reporting of this item should also be considered with care. Unlike with scales, this item cannot be assessed over time (e.g. change from baseline) or between groups since it is conditional on the actual use of concomittant medication. It is usually reported in a descriptive way, listing the proportion of patients per outcome category (e.g. 12% of the patients who took medication rated this as very helpful).

The QLQ-C30 is composed of both multi-item scales and single-item measures. These include five functional scales, three symptom scales, a global health status / QoL scale, and six single items. Each of the multi-item scales includes a different set of items - no item occurs in more than one scale.
All of the scales and single-item measures range in score from 0 to 100. A high scale score represents a higher response level. Thus a high score for a functional scale represents a high / healthy level of functioning, a high score for the global health status / QoL represents a high QoL (ie. a better state of the patient).
But a high score for a symptom scale / item represents a high level of symptomatology / problems (ie a worse state of the patient).
This information can be found in the QLQ-C30 Scoring manual. This manual is available to download via our website on the manuals page.

Unfortunately, this is not done in a verfied manner yet; but of course you can explore it. We are always interested in such results.

Yes, you can use single items from different modules. You will need to contact the Quality of Life Department prior to obtain authorization for such use. You should also acknowledge use of questionnaire items from the EORTC Quality of Life Group Item Bank in any publication. We would recommend that you explain in the methods section of your reports why you chose to use single items rather than the entire questionnaire module from which those items were extracted.

The order of the items cannot be changed. The questionnaire has been developed and validated in this format, and changing the order of the items might affect the psychometric properties of the questionnaire.

We recommend that additional items or questionnaires should be included after the QLQ-C30 and the chosen modules.

Any changes regarding modifications of the QLQ-C30 need to be approved by the executive committee of the EORTC Quality of Life Group and, in the case of questionnaire modules, by the primary module developer/author as well. The QLQ-C30 and modules are copyrighted instruments. For more information on this subject, you can contact Ken Cornelissen.

A short version of the questionnaire is not available. However, EORTC is considering this presently and may in the future develope a shorter version. In addition, a short form for use in palliative care, the QLQ-C15-PAL is developed, based on items from the QLQ-C30. Please see our modules page.

Statistical research carried out by Quality of Life Group members seems to suggest that a summary score is a poor representation of the data available in all the scales.

A number of research projects are underway to develop electronic versions of the QLQ-C30. However, you should remember that you will always need to complete a user's agreement.

As for the best way to analyze longitudinal data, there is no straight answer. The most important thing to keep in mind is that the analysis techniques should reflect your objective. This depends on the objective of the study, its design, the amount of missing data, sample size, available software, etc...

The most common approaches can be divided in two categories:

1. Summary measures.
   The idea of this approach is to summarize the repeated data (per patient) into one outcome. This one outcome can then be analysed using standard statistical techniques (t-test, chi-square-test, etc.).
   Common summary measures are:
   • analysis at each time point separately
   • area under the curve
   • change from baseline
   • time until event (predefined in/decrease)
   The problem with these type of analyses is that they ignore the longitudinal character of the data and therefore don't use the data to its full extent.
2. 2. Longitudinal modelling
   This approach will try to model the data, including the time dependence. Estimates of time and group parameters allow to test the significance of comparisons and or construct confidence intervals for predictions. Many different types of models are possible according to the assumptions one is willing to make. Most commonly used are general linear models (GLM) and general linear mixed models (GLMM).

Some good references are:

• Fairclough DL. Design and Analysis of Quality of Life Studies in Clinical Trials. Chapman and Hall/CRC Press, Boca Raton, Florida, 2002.
• Diggle, P. J., Liang, K. Y. & Zeger, S. L. Analysis of Longitudinal Data, Oxford University Press, 1994.
• Fahrmeir, Tutz. Multivariate Statistical Modelling using Generalized Linear Models Springer Statistics, New York, 1994.
• Verbeke, G. and Molenberghs, G. Linear Mixed Models for Longitudinal Data, New York: Springer - Verlag, 2000.

This depends on the research question you are aiming to answer with your project. The sample size should reflect the pre-specified objective, design and analysis.
In order to calculate the number of patients needed, the following parameters are required:

• type and number of tests or decision rules
• expected difference and variance
• significance level = Type I error = alpha
• power = 1 �| Type II error = 1- beta

The needed number of patients can then be calculated. In addition, the obtained number of patients needs to be corrected for possible missing data and/or attrition.

The actual number will depend on your study�fs specific objective, design and analysis techniques. As an example, suppose you want to detect a 10 point difference between two treatment arms. The endpoint is the physical functioning scale at a certain time-point with baseline correction. The expected variance for that scale is assumed to be 25 points. The chosen test is a two-sided t-test with alpha set to 5% and beta 20%. The needed number of patients is then 100 patients per arm. If we assume an attrition rate of 10% and a missing data rate of 20% then the actual needed number of patients to be entered is 139 patients per arm = 278 patients.

The QLQ-C30 and its modules have been designed to evaluate change of HRQoL in clinical trials setting. As such, a single individual score is not considered to be informative. Scores are only informative when used in a comparative setting:

• comparing different patient groups
• comparing changes within one group over time
• comparing changes over time between different patient groups.

When comparing scores, one should take into account that statistically significant differences do not necessarily imply clinically relevant differences and vice versa. For the QLQ-C30, a change in any scale of at least 10 points is considered to be clinically relevant (Ref: Osoba D, Rodrigues G, Myles J, Zee B, Pater J. Interpreting the significance of changes in health-related quality of life scores. J Clin Oncol 1998; 16 (1): 139-144).

Reference data are historical outcomes from the QLQ-C30 and/or its modules as obtained by previous clinical studies. Such data may be used for:

• comparison with a group of patients with similar characteristics. Ideally comparisons should be done within randomised trials, but this is not always feasible. If so, reference data provides a (limited) alternative
• investigate distribution of scores. Knowledge about the distribution of a certain scale for a certain patient group can help in the design of new studies
• sample size calculation. Reference data may be used to build assumptions needed for sample size calculations
• categorizing of scales. Sometimes categorised outcomes are preferred over the continuous scales. Reference data can help establish a priori justified cut-off point
• translation procedures. Comparison of data from new translation with available reference data can be helpful in spotting obvious discrepancies caused by incorrect/dubious translations.

Since the tables present the total sample size and the percentage in each group, you could easily calculate the mean and SD if you really want to. (Standard stats books tell how to do this for grouped data). But many of the scales are asymmetric (depending on the disease / stage grouping), many have floor or ceiling effects, and all are discrete not continuous. Mean and SD would not reflect this. E.g. for many scales the (mean) +/- (2xSD) will give "impossible" values outside 0 to 100.

That is why the reference values did not present mean and SD --- it is easy to do it if you want it, but we did not wish to encourage uncritical use of statistical methods.

To test significance, you can use for example a chi-squared test for trend (e.g. Altman DG. Practical Statistics for Medical Research, publ. Chapman & Hall, p 261).

References to publications on the questionnaires can be found in the bibliography.

If you have signed a user's agreement to use the module in your study, you may publish the data from the module.

No! Only the developers may publish the module on behalf of the EORTC Quality of Life Group.

No! We can only authorise people to use a Specimen of the questionnaire. A Specimen is a copy with a watermark across it. Our measures are protected by copyright, so we cannot allow unauthorised use of our questionnaires. If you want to receive a Specimen to put in a leaflet, please contact Ken Cornelissen. When you publish the Specimen, you also need to mention the full reference of the original publication of the validation paper of the QLQ-C30. Plus you need to add a statement that if persons want to use our measures, that they need to contact the Quality of Life Department. You can also mention our website address.

The ideal situation would be to have information on the patient's quality of life prior to their diagnosis of cancer. Since this is rarely available, the pre-treatment score is usually considered to be their starting point or baseline assessment. A pre-randomization assessment provides a starting point for assessing changes caused by both treatment and disease status. More information on this subject can be found in the 'Guidelines for Quality of Life Assessment' which can be obtained from Ken Cornelissen.

This is not possible, since the module has been designed to be used together with the core questionnaire, and the content validity is based upon this combination.

When Qol is used as an outcome measure in clinical trials, one (or some) subscales or single items should be selected as the main outcome measure, while the other will give additional descriptive data. Results can be reported as main scores, frequencies of patients on the various response categories and proportions of patients with high/low scores. We recommend that mean scores on all scales are reported in order to make easier to compare results between studies later.

In principle it is the patient him/herself who has to fill out quality of life forms and preferably without help from others. In the case where a patient is too sick to fill out the questionnaire or if the patient is not able to fill out the questionnaire for reasons such as forgetting his/her glasses, another person (preferably a hospital staff member) could read the questions and mark the patient's answers. This person should not make any suggestions to the patients but simply report the answers on the forms. If a patient received this type of help, it is usually asked to note this on the form.

It is generally agreed that patients are the best raters of their own quality of life. There are circumstances in which it is difficult or even impossible for patients to rate their own quality of life (e.g. patients who are cognitively impaired due to their cancer, terminally ill and children). In these circumstances their quality of life may be assessed by a third person, a so-called proxy. This can be a family member or the care giver. If it is anticipated that an increasing percentage of the patient population under study will be unable to complete questionnaires during the course of the trial, proxy respondents could be considered from the trial outset. More information on this subject is given in the 'Guidelines for Assessing Quality of Life in EORTC Clinical Trials'.

In principle it is the patient him/herself who has to fill out quality of life forms and preferably without help from others. Some patients may need help, e.g. because of reduced sight. Another person could then read the questions, without making any suggestions to the patient, and report the answers on the form. Some patients may ask for advice- What do you think?"- but manage to answer themselves when they are asked again.

At entry in a study, the patient should be given an explanation of the objective of the study and instructions for filling out questionnaires. The following issues should be explained to the patient:

• The schedule of assessments.
• The questionnaire is a self-administered questionnaire that should be filled out preferably by the patient him/herself.
• The patient should circle the choice that best corresponds to his/her situation.
• There is no right or wrong answer to any of these questions.
• All questions should be answered.

Most elderly patients can cope with the questionnaires if the scoring system is explained to them.

The majority of people complete the questionnaires within 10 to 15 minutes. However, patients should be given the time they need to complete the questionnaire.

Basically, as long as it is the patient that is being interviewed and the person doing the interview with the patient does not influence their answer, there should not be a problem doing it this way.

Actually no. It's not developed or validated on children. You need to search a special tool that is done this way. EORTC doesn't have such tools as we don't do much QOL in this age group. We don't know where to recommend for this, but you can search the literature/pubmed for the most used tools or seek local experts.

No, this information is not needed per se. It is not part of the actual HRQoL outcomes. This question may be omitted or modified. However, be aware that in most studies one will need an unambiguous method of matching a person's questionnaire with other relevant study data. Many regulatory bodies (e.g. The World Medical Association�s Declaration of Helsinki, International Conference on Harmonization (ICH), Good Clinical Practice (GCP) and EU Clinical Trials Directive guidelines) require that a study subject�s right to data protection and privacy must be safeguarded in the context of a clinical study. This usually implies that the full names of the study participants cannot be collected since these must remain anonymous to the study investigators handling the data. If you are unsure which privacy data you can or cannot collect, please check with your local or national governing Ethical Committee for guidance.

If you want to attend a meeting, you can contact the Group's Secretary here. Conditions to become a member of the Group can be found here.

The EORTC QLQ-C30 has indeed been used in several registration studies. However, the FDA doesn't generally base their opinion on quality of life data. For more information you can always contact the FDA, or the Head of the Quality of Life Department.

PROQOLID: Patient-Reported Outcome and Quality of Life Instruments Database

Developed by Mapi Research Institute and managed by Mapi Research Trust. PROQOLID aims to identify and describe QOL instruments to help you choose appropriate instruments and facilitate your access to them. www.qolid.org.

OLGA (The On-Line guide to Quality of Life Assessment)
A database with decision-theoretic selection algorithms to help in the selection of instruments. www.olga-qol.com.

Maurice J. Staquet, Ron D. Hays, Peter M. Fayers. Quality of Life Assessment in Clinical Trials. Oxford University Press, 1998.

And you can also consult our links page for more information.