Abstract
The EuroQol Group is developing a new EQ-5D-Y-5L version with 5 severity levels for each of the 5 dimensions. The 5 severity levels describe different health severities and there is a potential for severity level inversion. This article aims to report the process of cross-cultural adaptation of the beta EQ-5D-Y-5L into Chichewa (Malawi) using the card ranking exercise, which has been added to the EQ-5D-Y-5L translation protocol.
To assess the correct hierarchical ordering of severity levels, the adaptation followed the EQ-5D-Y-5L translation protocol. Cognitive interviews were undertaken to establish conceptual equivalence. Thereafter, 4 iterations of ranking exercises were conducted, leading to amendments of the translated Chichewa version to arrive at a final version.
The iterations were assessed by 18 participants aged 8 to 14 years. Health proved to be a difficult concept to translate as was “discomfort.” Cognitive interviews identified further conceptual issues, particularly with the “looking after myself” dimension. Considerations about lack of soap or water indicated that some children did not fully comprehend this dimension as being about the ability to wash and dress themselves. The iterative card ranking exercise detected severity level inversion between “a little bit” and “some,” and between “a lot” and “extreme” and alternative Chichewa words/phrases were then tested. Ultimately, the intended hierarchical severity ranking was achieved and an acceptable Chichewa version was produced.
Conceptual and linguistic equivalence to the English EQ-5D-Y-5L was established for the Chichewa EQ-5D-Y-5L version. The card ranking exercise was instrumental in correcting severity level inversion and supporting the comprehensible translation.
•The EuroQol’s Version Management Committee has a standard cross-cultural translation protocol for all of the EQ-5D versions (EQ-5D-Y, EQ-5D-3L, EQ-5D-5L). The protocol aims to establish both conceptual and linguistic equivalence of the EQ-5D versions cross-culturally. A ranking exercise has now been incorporated into the cognitive debriefing.•There were items of the new beta EQ-5D-Y-5L for which linguistic and conceptual equivalent terms were difficult to identify in Chichewa. Additional descriptors needed to be tested to ensure all concepts were understood by children in Chichewa. Severity level inversion is a major challenge for a health-related quality of life questionnaire with 5 severity levels, particularly in children. This article describes how the potential for level inversion can be decreased during the cross-cultural adaptation of the EQ-5D-Y-5L.•Instruments developed in high-income, mostly English-speaking contexts need to be interrogated in-depth for use in middle to low-income countries. Care needs to be taken in health-related quality of life instruments with multiple severity levels, such as the beta EQ-5D-Y-5L, that no severity level inversion is present in the descriptors. This is particularly important if preference-based weights are to be developed for the use of the instrument in economic evaluation and health technology assessments. The resolution of severity level inversion as reported in this article is a necessary first step toward attaining this during the translation phase.