Rotterdam, The Netherlands – The Clinical COPD Questionnaire (CCQ) is a widely-used questionnaire to assess the effectiveness of chronic obstructive pulmonary disease (COPD) treatments. However, it is not a preference-based measure, and hence cannot be directly used to estimate the cost per quality adjusted life year (QALY) gained. Since many clinical trials include CCQ rather than preference-based measures, a successful mapping of the CCQ to the preference-based EuroQoL-5D (EQ-5D) would facilitate performing economic evaluations with COPD trial data.
In a new study by the Institute for Medical Technology Assessment (iMTA) of Erasmus University, Rotterdam, a model was developed to predict EQ-5D values from CCQ data. The researchers obtained data from three trials including 5,751 observations with a broad range of COPD severity. Three approaches were used to estimate the statistical relationship between CCQ and EQ-5D in a development set: ordinary least squares, generalized linear models, and TOBIT regression. Performance was tested in the development data set and in external trial data sets. The best performing model was selected based on the mean absolute error and root mean square error of predicted versus observed values in both the development data set and external data sets. The researchers also developed models for different country-specific EQ-5D value sets.
“The models predicted mean EQ-5D values that were similar to the observed mean values in the development set, but the EQ-5D values did significantly differ in external data sets. The mapping models underestimated the EQ-5D values for the mild health states, while they overestimated them in more severe health states. Therefore, the current model is only applicable in data sets that are similar to the development set. In general, mapping models should be used with caution.” said author Melinde Boland.
The full study, “Mapping the Clinical Chronic Obstructive Pulmonary Disease Questionnaire onto Generic Preference-Based EQ-5D Values,” is published in Value in Health.