Construct and convergent validity and repeatability of the Questionnaire d'Activité Physique pour les Personnes Âgées (QAPPA), a physical activity questionnaire for the elderly.

ABSTRACT

OBJECTIVES: To investigate repeatability, and construct and convergent validity of the Questionnaire d'Activité Physique pour les Personnes Âgées (QAPPA), a self-reported physical activity questionnaire for the elderly. STUDY DESIGN: Cross-sectional and prospective (1-year follow-up) data from two postal surveys among members of the medical insurance scheme of the French national education system. METHODS: A questionnaire on general health and the QAPPA were sent out twice, with a 1-year interval. Participants were ≥60 years of age (sample size varied from 225 to 393 according to the analysis). Intraclass correlations, Bland-Altman plots (continuous variables) and kappa coefficients (categorical variable) were used for repeatability analyses. Comparison tests were performed (Wilcoxon rank sum test and Kruskal-Wallis test) for construct validity. Chi-squared test and correlation coefficients were used to establish the convergent validity of the QAPPA. RESULTS: The QAPPA had acceptable repeatability with a 1-year interval, with moderate test-retest correlation coefficients (from r = 0.46 to r = 0.64) and a fair-to-good kappa coefficient (κ = 0.44). It discriminated between individuals by age group (with younger persons being more physically active than older persons) and exercise behaviour (with regular exercisers being more active than non-exercisers). The QAPPA was negatively associated with decline in physical function and comorbidity, and was positively associated with self-reported health and satisfaction with body functioning. CONCLUSION: This study found that the QAPPA is a valid tool for assessing the physical activity of older adults. It could be used for research purposes as well as for epidemiological surveillance.