Comparison of the IPAQ- Short Form and Accelerometry Predictions of Physical Activity in Older Adults

Abstract

Listen

Translate article

PURPOSE: To compare physical activity (PA) levels obtained from the International Physical Activity Questionnaire (IPAQ) short form and accelerometers (ACC) in a group of 50-87 year old adults. METHODS: One hundred and forty-one adults (36 men, 66.3 ± 8.7 y; 105 women, 63.5 ± 7.3 y) participated in this study. Each participant wore an ACC (Actigraph 7164, Fort Walton Beach, FL) for 7 consecutive days during all waking hours. ACC derived PA was categorized into time spent in inactivity (≤260 cts/min), moderate (760-5725 cts/min), walking (1952-5725 cts/min), vigorous (≥5725 cts/min), and combined moderate and vigorous PA (MVPA) (≥760 cts/min). Participants completed the IPAQ which assesses time spent in inactivity (sitting), moderate, walking, and vigorous PA. Total ACC MVPA and total IPAQ activity were separated into PA tertiles. The sample was further separated into either met or did not meet PA recommendations (recs) (30 min moderate PA ≥5 days/week and/or 20 min vigorous activity ≥3 days/week). Pearson correlations were used to assess relations between time spent in PA. Percent (%) agreement and kappa statistics were used to compare PA tertiles and those who met or did not meet recs. RESULTS: Comparisons across PA measures showed significant relationships between inactivity and sitting time stratified by gender and for the whole group (r=0.29-0.38, p<0.01). Time spent walking (r=0.29-0.32, p<0.01) and time spent in vigorous activity (r=0.23, p<0.02) were significantly correlated in females and the overall group only. Non-significant findings were observed in moderate activity for all groups (r=0.11-0.13, p=0.19-0.44). MVPA and IPAQ total activity were correlated for the total group only (r=0.29, p<0.01). Percent agreement between PA categories across PA measures were 53% for males (K=0.29, p= 0.01), 30% for females (K= −0.05, p= 0.47), and 36% for the whole group (K= 0.04, p= 0.50). The IPAQ signified 75% as meeting PA recs, whereas, the accelerometer signified 5% as meeting PA recs. CONCLUSIONS: Overall findings are not dissimilar to those reported for other self report PA questionnaires, demonstrating moderate agreement between self report PA and objectively determined PA. Due to the drastic differences in estimating prevalence of PA recs, the IPAQ may not be a good measure of such in this older population.