Abstract

As both a user of translated measures and a scientist interested in moving the field forward, I am often left after a translation project with a set of uneasy questions about the translation process. In spite of the care taken during the translation, have we really found the equivalent construct in the target language and culture? Do people who speak Spanish and live in Mexico really understand health-related concepts differently from those in Argentina, that is, do we really need to go through all those complicated steps of adaptation? How many people will actually fill out the scale in the language once it is translated (almost certainly never enough to support a full psychometric validation)? If we have a reasonably good translation such as is provided by a forward translation and reconciliation only, how much more value and quality are added by the back translation and cognitive debriefing steps, that is, is the cost of performing those additional steps justified? All of these examples are not abstract, but are derived from specific experiences over the years that have arisen in translating outcomes measures. Even though the article by Wild et al. in this current issue of Value in Health presenting the report of the ISPOR Task Force on Translation and Adaptation was at least partly aimed at laying such questions to rest by putting out guidelines, I hope that Wild’s article and the accompanying editorial by McKenna and Doward (also in this issue), will actually stimulate greater debate and empiric research on the thorny problems of translation.