The Sensitivity And Specificity Of The IPAQ For Detecting Intervention Related Changes In Physical Activity

Abstract

Many intervention studies rely on self-reports of physical activity because they are more feasible than objective measures and the type and domain of physical activity (PA) can be measured. However, few data are available that examine the sensitivity and specificity of self-report measures of PA for detecting intervention-related change. PURPOSE: to examine the sensitivity and specificity of the IPAQ for detecting intervention-related changes in PA compared to accelerometer measurement, in a community sample of Latinas. METHODS: A community sample of 94 women (mean age = 36.31 ± 9.1 yr; mean body mass index [BMI] = 31.37 ± 7.13) participated in a 12 week pedometer-based program to increase the number of minutes engaged in moderate intensity physical activity (MPA) during leisure time. Participants completed the Spanish-language IPAQ (telephone, long form) and wore an ActiGraph accelerometer for seven days. RESULTS: Significant increases in MPA were detected by the IPAQ (Medianbaseline = 435 min. vs. M post-intervention = 667.5 min.; +50% over baseline) and the accelerometer (Mbaseline = 89.81 min. vs. M post-intervention = 154.56; +68% over baseline). The difference in magnitude was not statistically significant. An increase of 333% in IPAQ measured walking for leisure was reported (Mbaseline = 7.5 min. vs. M post-intervention = 60 min.). However, there was no statistically significant relationship between IPAQ-measured change in PA and accelerometer-measured change in PA. Based on accelerometer data, IPAQ sensitivity and specificity at baseline was 81% and 30%, respectively, and 85% and 11% at post-intervention. CONCLUSION: These data suggest that the IPAQ is sensitive to change in PA but changes are largely uncorrelated with objectively measured changes in a sample of community dwelling Latinas. The specificity of the IPAQ for correctly identifying Latinas who do not meet PA guidelines is poor.