Validation of the Physical Activity Scale for individuals with physical disabilities.

OBJECTIVE To determine the criterion validity of the Physical Activity Scale for Individuals With Physical Disabilities (PASIPD) by means of daily physical activity levels measured by using a validated accelerometry-based activity monitor in a large group of persons with a physical disability. DESIGN Cross-sectional. SETTING Participants' home environment. PARTICIPANTS Ambulatory and nonambulatory persons with cerebral palsy, meningomyelocele, or spinal cord injury (N=124). INTERVENTIONS Not applicable. MAIN OUTCOME MEASURES Self-reported physical activity level measured by using the PASIPD, a 2-day recall questionnaire, was correlated to objectively measured physical activity level measured by using a validated accelerometry-based activity monitor. RESULTS Significant Spearman correlation coefficients between the PASIPD and activity monitor outcome measures ranged from .22 to .37. The PASIPD overestimated the duration of physical activity measured by using the activity monitor (mean ± SD, 3.9±2.9 vs 1.5±0.9h/d; P<.01). Significant correlation (ρ=-.74; P<.01) was found between average number of hours of physical activity per day measured by using the 2 methods and difference in hours between methods. This indicates larger overestimation for persons with higher activity levels. CONCLUSIONS The PASIPD correlated poorly with objective measurements using an accelerometry-based activity monitor in people with a physical disability. However, similar low correlations between objective and subjective activity measurements have been found in the general population. Users of the PASIPD should be cautious about overestimating physical activity levels.