Reliability and validity of two multidimensional self-reported physical activity questionnaires in people with chronic low back pain

Abstract

Background Although there is some evidence for reliability and validity of self-report physical activity (PA) questionnaires in the general adult population, it is unclear whether we can assume similar measurement properties in people with chronic low back pain (LBP). Objective To determine the test-retest reliability of the International Physical Activity Questionnaire (IPAQ) long-version and the Baecke Physical Activity Questionnaire (BPAQ) and their criterion-related validity against data derived from accelerometers in patients with chronic LBP. Design Cross-sectional study. Methods Patients with non-specific chronic LBP were recruited. Each participant attended the clinic twice (one week interval) and completed self-report PA. Accelerometer measures >7 days included time spent in moderate-and-vigorous physical activity, steps/day, counts/minute, and vector magnitude counts/minute. Intraclass Correlation Coefficients (ICC) and Bland and Altman method were used to determine reliability and spearman rho correlation were used for criterion-related validity. Results A total of 73 patients were included in our analyses. The reliability analyses revealed that the BPAQ and its subscales have moderate to excellent reliability (ICC2,1: 0.61 to 0.81), whereas IPAQ and most IPAQ domains (except walking) showed poor reliability (ICC2,1: 0.20 to 0.40). The Bland and Altman method revealed larger discrepancies for the IPAQ. For the validity analysis, questionnaire and accelerometer measures showed at best fair correlation (rho < 0.37). Conclusions Although the BPAQ showed better reliability than the IPAQ long-version, both questionnaires did not demonstrate acceptable validity against accelerometer data. These findings suggest that questionnaire and accelerometer PA measures should not be used interchangeably in this population.