The present study investigated the convergent validity of an interview-administered IPAQ long version (IPAQ-L) in an older population by comparison with objective accelerometry movement data. Data from 52 participants (mean age 67.9 years, 62% male) were included in the analysis. Treadmill derived (TM-ACC: 1,952-5,724 cpm) and free-living physical activity (PA) derived (FL-ACC: 760-5,724 cpm) accelerometer cut-points were used as criterion. IPAQ-L measures (total PA, leisure-time, walking-time, sedentary time) were significantly correlated with accelerometry (P ≤ .05). Differences in sex were observed. Bland-Altman Limits of Agreement analysis showed that the IPAQ-L overestimated PA in relation to accelerometry. Our results show that an interview-administered IPAQ-L shows low to moderate convergent validity with objective PA measures in this population but there may be differences between males and females which should be further investigated.
Keywords: International Physical Activity Questionnaire; cancer; elderly; measurement; older adults; physical activity.