Validity Of The International Physical Activity Questionnaire - Short Form Among Individuals With Visual Impairments

PURPOSE: The International Physical Activity Questionnaire - Short Form (IPAQ-SF) is regularly used to assess physical activity (PA) and sedentary time (SED) in visually impaired populations. However, little is known about the internal validity of the IPAQ-SF in this population and if validity differs based vision level. The purpose of this study was to assess validity of the IPAQ-SF among individuals with no usable vision (NUV) and some usable vision (SUV). METHODS: Visually impaired participants (total n = 81) with either NUV (n = 57) or SUV (n = 28), wore a waist-mounted GT3X+ accelerometer for 7 days during all waking hours. Upon returning the device, participants completed the IPAQ-SF. Self-reported moderate-, vigorous-, and moderate-to-vigorous PA (MPA, VPA, MVPA, respectively) and SED were compared with that from the GT3X+ using Spearman’s rho for the group overall and by vision level. Mean absolute error was calculated for IPAQ-SF variables using GT3X+ estimates as the comparison. RESULTS: IPAQ-SF estimates of MVPA, MPA, VPA, and SED did not differ between participants with NUV and SUV (all p > 0.05). The GT3X+ data suggested that those with NUV showed higher amounts of SED compared to those with SUV (p < 0.05) but no differences were detected in PA of different intensities. Concurrent validity of MVPA and MPA between IPAQ-SF and GT3X+ was moderate (range: 0.410 - 0.610; all p < 0.05) for the group overall, and when separated into NUV and SUV groups, as well as for VPA in the SUV group. Concurrent validity of VPA was poor overall and moderate in the SUV group but statistically significant in both instances (0.303 and 0.410, respectively; p < 0.05). Concurrent validity for SED was poor overall and in both groups (-0.227 - 0.221 all p > 0.05). For all variables, MAE was higher in the SUV group than the NUV group. CONCLUSIONS: The validity of estimates of MVPA and MPA from the IPAQ-SF appear to be strong for individuals with NUV and SUV and similar to those in the general population. Researchers should use caution when attempting to assess VPA among those with NUV via IPAQ-SF. Alternative options for assessing SED in this population are recommended as the validity of estimates from IPAQ-SF are lower than estimates for the general population.