Response to: A Methodology for Successfully Producing Global Translations of Patient-Reported Outcomes Measures for Use in Multiple Countries by Two et al.


  • Tamzin Furtado BA (hons)

To the Editor—I am writing regarding the recent Value in Health article concerning the issue of creating one language version of a patient-reported outcome measure for use in multiple countries (for example, one French version of a questionnaire that is suitable for use in Canada, France, and Belgium) [1].

The 2009 International Society for Pharmacoeconomics and Outcomes Research (ISPOR) Task Force article addresses this issue in-depth, concluding that there is no “right” way to perform the procedure but that it depends on the measure, countries, and available timelines and budget of the individual project. It was interesting therefore to hear the procedure proposed by the authors of having a teleconference between each step of the harmonization procedure heralded as creating a “superior final translation.”

In practice, it may be that the methodology is somewhat cumbersome in comparison with some alternatives. Also the problem as discussed in ISPOR's 2009 Task Force report still remains, that any translation achieved will be a compromise, no matter how thorough or rigorous the teleconference at each step is; it may be that the wording used is simply the “best available alternative” for the countries involved and not the best wording for each individual country.

Although an alternative procedure appears commendably thorough, it may be that the extra time and cost involved with multiple teleconferences (which heavily influence the budget of a project) does not result in a translation which is superior to that of an alternative methodology for creating a harmonized translation, which would be for a lead investigator (for example, the French for France in-country language consultant might be chosen as the “lead”) to create the reconciled translation and reply to back translation queries, with suggestions then discussed with the other investigators via e-mail. This method is therefore less costly for the sponsors involved and still results in the reduction of “risk of influence of varying word structures” as the authors mention.

It would be interesting to further compare the translated results of the methodologies available to determine the merits of this method.—Tamzin Furtado, BA (hons), Oxford Outcomes Ltd., Oxford, UK.

Source of financial support: None.