To adapt the DUKE Health Profile, a 17-item self-report generic health related quality of life measure cross-culturally in french. A multidisciplinary expert committee was provided with three translations independent of each other, each backtranslated to the original language, and produced a synthesis version equivalent to the original. A cohort of 963 persons from the general population filled in the questionnaire twice in three months. The internal consistency was acceptable (Cronbach's alpha = 0.63-0.81) except in social dimension. Convergent validity was evidenced by a significant correlation with overall health. The test-retest reproducibility in stable subjects (601) was satisfactory (intraclass coefficient correlation r = 0.63-0.78) except in pain and disability dimensions. There was a significant modification of scores in the same direction as overall health change in subjects improved (n = 128) or worsened (n = 187). Age-adjusted scores were lower in females, in subjects with lower education, urban residency, unemployed and living alone and in subjects reporting a chronic disease. This short form questionnaire similar to the original version proved simple to use in the general population.