French translation and cultural adaptation of a questionnaire for patients with hip or knee prosthesis

  • Erivan Roger
  • Villatte Guillaume
  • Chaput Thibault
  • Mulliez Aurélien
  • Ollivier Matthieu
  • Descamps Stéphane
  • Boisgard Stéphane

  • Orthopedics
  • Patient follow-up
  • Questionnaire
  • Translation
  • Total hip arthroplasty
  • Total knee arthroplasty


Introduction The Kingsbury questionnaire offers the possibility of follow-up by means of an X-ray and a simple questionnaire sent to the home address of the patient, who will not need to come in consultation if there are no problems. The questionnaire detects any anomaly in follow-up. In case of anomaly in the questionnaire or radiograph, the patient is contacted and/or seen again, as appropriate. The Kingsbury questionnaire has had no transcultural validation, and we therefore conducted a prospective study in order to 1) translate into French the questionnaire, previously validated in English; 2) adapt it for good understanding according to cultural habits; and 3) assess the translated version on a test-retest procedure. Hypothesis The study hypothesis was that the translated questionnaire would show good test-retest reproducibility. Material and methods The exact English version of the questionnaire was obtained directly from the authors of the index publication. A methodology of translation, back-translation and test-retest enabled assessment of the translation and of the reproducibility of the French version. The reference method of cultural adaptation of self-administered questionnaires and patient information documents was used. The questionnaire was tested prospectively. Results One hundred patients were contacted, providing 73 clinical tests with radiographic validation and 48 complete test-retests in a representative population of total hip and total knee arthroplasty (THA, TKA). Internal coherence showed a KR-20 coefficient of 0.71 and Cronbach alpha of 0.76: e.g., good internal coherence. Item difficulty, requiring renewed contact, was low for all questions. Mean variance was low on the first 7 questions: 0.08 (range, 0.02–0.16). Correlation was close to 0.5 for each question. Analysis of reproducibility found excellent agreement (> 90%) for the first 7 questions, which were binary; for question 8, agreement was good (83.3%) considering that there were 5 possible responses. For 19 of the 73 respondents, the questionnaire results indicated a need for further contact. After analysis of their radiographs, 4 needed to be seen in consultation again. The other 15 had unfavorable responses but without deterioration since the last classical consultation or any radiologic abnormality consultation, and were not called back for consultation. Conclusion The French version of the Kingsbury questionnaire provided reproducible assessment, avoiding the need to call the patient back for consultation unnecessarily. The questionnaire needs validating in a larger sample before being widely used: the present study was just a first step.