Validation of two questionnaires to assess the level of physical activity and sedentary time in a Colombian university community

Abstract

Objective: Evaluate the concurrent validity and test-retest reliability of the Global Physical Activity Questionnaire (GPAQ) and the International Physical Activity Questionnaire (IPAQ, short version) in a university community. Methodology: A total of 76 people from the University of Antioquia, Medellín (Colombia), aged between 18 and 44, carried an accelerometer (ActiGraph GT3X +) for seven days and subsequently filled out the two questionnaires; these were repeated a week later. The validity and reliability for assessing energy expenditure (EE) and sedentary time (ST) was determined with the intraclass correlation coefficient (ICC); concordance to classify the physical activity level (PAL) was evaluated with the weighted Kappa index (wKI). Results: To measure the EE, IPAQ showed a “moderate” validity (ICC = 0.434, CI 95% = 0.231-0.600; p <0.001) and the “poor” QPAQ (ICC = 0.335, CI 95% = 0.123-0.519; p <0.001); for ST, the validity of both questionnaires was "poor" (IPAQ: ICC = 0.234, CI 95% = 0.011-0.435, p <0.020; GPAQ: ICC = 0.231, CI 95% = 0.009-0.432, p <0.021). To classify the PAL (low, moderate or high), both questionnaires showed "low" concordance (IPAQ: wKI = 0.298, p = 0.001; GQAP: wKI = 0.137, p = 0.119). The reliability of the GPAQ for assessing EE was “good/excellent” (ICC = 0.838; CI 95% = 0.754-0.895; p <0.001) and the IPAQ was “moderate” (ICC = 0.655; IC 95% = 0.505-0.766; p <0.001). The reliability for ST was "moderate" in both instruments (IPAQ: ICC = 0.716; CI 95% = 0.583-0.811; p <0.001; GPAQ: ICC = 0.736; CI 95% = 0.613-0.824; p <0.001). Conclusions: The questionnaires showed little validity for measuring PAL and ST . However, both instruments presented good reliability for measuring the variables studied. The use of questionnaires for the population studied is not recommended.