The original PPAQ authors previously assigned each activity on the PPAQ to a specific metabolic equivalent (MET) value using field-based measurements in pregnant women (17) and the 2000 update to the compendium-based MET values (1 MET = 1 kcal/kg x hour).
We compared the PPAQ results to the Actical[R]-measured PA in the combined sample of all women and in the active and nonactive women separately.
Paired sample t-tests and Wilcoxon signed-rank tests were used to test for potential differences in the PA variables between the PPAQ and the Actical[R] data.
The time spent in different intensities of PA (sedentary, light PA (LPA), moderate PA (MPA), vigorous PA (VPA), and MVPA) as measured by the PPAQ and the Actical[R] are shown in Table 2.
The comparison of the PPAQ to the Actical[R] when investigating the active and the non-active women separately is shown in Table 4.
PPAQ variables did not correlate with their respective Actical[R] measures for sedentary time, LPA, MPA, VPA or MVPA (Table 5).
Our data indicate that the PPAQ is not an accurate proxy of PA: it drastically overestimates LPA, MPA and MVPA in all women, and PPAQ-measured PA variables did not correlate with the Actical[R] measures of sedentary time, LPA, MPA, VPA or MVPA.
These results conflict with the previous publications that concluded that the PPAQ was a reasonably accurate tool for assessing prenatal PA.
Using the PPAQ and misclassifying a number of the non-active women as active is problematic, especially if this questionnaire were to be employed in research exploring the health effects of prenatal PA.
This gross overestimation of PA and the lack of correlation between the PPAQ and the Actical[R]-measured PA call into question the findings from the recently published studies that used the PPAQ to measure PA during pregnancy.