A pragmatic approach ensuring accuracy in language translation in tuberculosis research

Background: To research patient perceptions of medicines and illness in a multicultural setting, it is appropriate to translate research materials. However, the translation procedures should be valid and reliable to assure accuracy. Objectives: To translate into 3 languages-Turkish, Urdu, and Bengali-a research questionnaire investigating illness and medicines perceptions of tuberculosis patients and to validate the translation. Methods: A 4-stage protocol for the translation and validation of research questionnaires investigating illness and medicines perceptions was designed and implemented. This involved forward and back-translation, group-validation, and post hoc conceptual equivalence rating in 3 different languages. Results: The translation protocol was found to be very useful in identifying discrepancies between original and translated versions; a total of 83 amendments were required. Post hoc evaluations also demonstrated improvements for 2 of the 3 language translations. Some redundancy was apparent and an improved protocol was suggested by the authors. Conclusions: This study demonstrated that including a protocol for translation and validation of the translation is crucial to assure accuracy in multicultural research.