Development and validation of a Spanish translation of the Yale activity questionnaire

Valid measures of physical activity are critical research tools. The objective of this study was to develop a Spanish translation of the Yale Physical Activity Survey, and to provide preliminary evidence of its validity in a population of Dominican patients with lower extremity arthritis.

A Dominican bilingual health care professional translated the Yale Physical Activity Survey (YPAS) from English to Spanish. Several Dominican adults reviewed the translation to ensure it was linguistically and culturally appropriate. The questionnaire was back-translated to English by a North American researcher who is fluent in Spanish. Discrepancies between the original and back-translated versions were resolved by the translator and back-translator. The Spanish translation was administered to 108 Dominican subjects with advanced hip or knee arthritis prior to (N = 44) or one to four years following (N = 64) total joint replacement. We assessed construct validity by examining the association of YPAS scores and measures of functional status and pain (WOMAC), quality of life (EQ-5D) and the number of painful lower extremity joints.

A higher YPAS Part II Activity Dimensions Summary Index score had weak to modest correlations with worse function and quality of life as measured with the WOMAC function scale (r = 0.21, p = 0.03), SF-36 Physical Activity Scale (r = 0.29, p = 0.004) and EQ-5D (r = 0.34, p = 0.0007). Total minutes of vigorous activity and walking had weak to modest correlation with these measures (WOMAC Function Scale (r = 0.15, p = 0.15), SF-36 Physical Activity Scale (r = 0.21, p = 0.04) and EQ-5D utility (r = 0.24, p = 0.02)). Correlations between the YPAS Part I energy expenditure score and these measures were lower (WOMAC Function Scale (r = 0.07, p = 0.49), SF-36 Physical Activity Scale (r = 0.03, p = 0.74) and EQ-5D utility (r = 0.18, p = 0.07)).

We have developed a new Spanish translation of the Yale Physical Activity Survey and provided evidence of convergent validity in a sample of Dominican patients prior to or 1–4 years following total joint replacement.