Normative Data for EORTC CAT

Patient-reported outcomes, i.e. outcomes that are directly obtained from the patient, are traditionally assessed using a static approach. That is, the same questions (items) are presented to each respondent regardless of their relevance to the individual. As a result, respondents often have to answer items that are irrelevant to them which increases respondent burden and may result in less precise score estimates. Collecting data via computerized adaptive testing (CAT) can address this shortcoming as it avoids the administration of irrelevant items by tailoring the instrument to the individual. To make use of the advantages of CAT, the EORTC Quality of Life Group has developed the EORTC-CAT which measures the same dimensions as the EORTC QLQ-C30 but with increased precision and flexibility. To enhance the interpretation of scores obtained via the EORTC-CAT, it is important to link the CAT to reference data obtained from the general population. Therefore, to increase the interpretability of the EORTC-CAT, the current project was aimed at collecting representative data of a European reference population for norming purposes with a final sample size of N=15,386. The following countries were included in our study: Austria*, Canada, Denmark*, France*, Germany*, Hungary*, Italy*, Netherlands*, Poland*, Russia, Spain*, Sweden*, Turkey, United Kingdom*, and USA.

(Countries marked with * form the European Norm.)

Data collection was finalized in April 2017 across 13 European countries, the USA and Canada. All data have been pooled into one large database where they were cleaned and weighted. They are now ready to be used.

In addition to the core papers published in 2019, we are planning several additional publications, in particular at the country level. Also, where available we aim to compare the general population data with patient data.

The computerized adaptive test of the EORTC (CAT) has the advantage that patients do not need to answer unnecessary questions. However, to understand the data we collect it is useful to obtain norm data from the general population. This is the aim of this project. With these norm data we can then compare a patient’s score relative to a healthy person matched by age and sex.