An assessment of the external validity of mapping QLQ-C30 to EQ-5D preferences

Although cancer-specific Health-related Quality-of-Life measures are commonly included in randomized clinical trials or other
prospective non-randomized clinical studies, it is rare that preference–based instruments are used, which allow the calculation of a Utility weight suitable for estimating Quality-adjusted Life-Years gained.

To test the external validity of a previously published mapping algorithm to transform the EORTC QLQ-C30 questionnaire responses
into EQ-5D-derived utilities by predicting EQ-5D utilities from QLQ-C30 scores.

Comparative retrospective data analysis of four multicentre, prospective clinical trials in Breast, Multiple Myeloma, Non-Hodgkin
Lymphoma and Non-Small-Cell Lung cancer patients with, respectively, 219, 172, 132 and 172 patients. Regression analysis of
individual pairs of EQ-5D and QLQ-C30 scores.

Although the internal predictive power of a previously published mapping equation was high, its external validity when tested
on a set of unrelated external data sets in other cancers proved to underestimate both the mean and variance of the mapped
EQ-5D utilities. Furthermore, it appears that the relationship between QLQ-C30 scores and EQ-5D values is not stable across
the different data sets.

Validation of the proposed algorithm in other external clinical data sets should be encouraged as well as the application
of other more complex mapping methods to enhance accuracy of mapping. In the meanwhile, direct mapping from QLQ-C30 profiles
to EQ-5D utilities using published algorithms should be performed with reservations.