Reproducibility and sensitivity of a physical activity questionnaire in Mexican people.

Objective. To assess the reproducibility and sensitivity of a physical acti vity questionnaire (PAQ) developed at Laval University, to detect differenc es in lean and obese individuals. Material and Methods. A cross-sectional s tudy was conducted at Mexico's National Institute of Medical Sciences and N utrition, between January and May 1999. The PAQ was translated into Spanish and adjusted to the Mexican setting. The test-retest method was used to me asure reliability, allowing a four-week interval between tests (n=30 overwe ight subjects). To assess the questionnaire's sensitivity a group of young cadets (n=18) was compared to a group of young civilians (n=32). Concordanc e was measured through the intraclass correlation coefficient; test-retest was analyzed using the paired or unpaired Student's t test, as necessary. R esults. The intraclass correlation coefficient was 0.86. The PAQ was able t o show differences of more than 400 kcal/day (1674 kJ/day) and more than 4 kcal/kg/day (17 kJ/kg/day), among young people with high physical activity (t-test). Conclusions. The PAQ is a sensitive and reliable instrument that can be used in the Mexican population. The English version of this paper is available at: http://www.insp.mx/salud/index.html.