Responsiveness of the single item measure to detect change in physical activity
journal contributionposted on 01.04.2022, 01:04 authored by Paul O'HalloranPaul O'Halloran, Michael KingsleyMichael Kingsley, Matthew NicholsonMatthew Nicholson, Kiera StaleyKiera Staley, Erica RandleErica Randle, Annemarie Wright, Adrian Bauman
Objective The objective of this study was to evaluate the responsiveness of the single item measure (SIM) to detect change in PA when compared to hip-worn accelerometry. A secondary aim was to provide further data on validity of the measure at a single time point. Methods Validity of the SIM to determine the number of days of ≥30 minutes of accelerometer-derived moderate-to-vigorous physical activity (MVPA) was assessed in 120 participants (78% female, 19% male, 3% other; age: 46 ± 11 years). Analysis of change was based on change in the number of days that participants completed ≥30 minutes of MVPA determined by the SIM and accelerometry over two 7-day periods in 90 participants (age: 47±11 years). Accelerometer data were analysed as total minutes of MVPA per day (MVPA-total) and as sustained bouts of 10 minutes or more of MVPA (MVPA-bouts). Validity of the SIM to detect change in MVPA, using accelerometer data as the reference measurement, was examined through Spearman’s correlation and agreement in classification of change between SIM and accelerometry. Responsiveness to change was assessed by standardised response means and Cohen’s d. Results Standardised response means for PA change were moderate for the SIM (0.77), MVPA-total (0.57) and MVPA-bouts (0.79). The correlation for change in number of days ≥30 minutes MVPA between the SIM and accelerometry were small to moderate (MVPA-total: r = 0.36 and MVP-bouts: r = 0.40). The SIM displayed moderate accuracy (60%-63%), using accelerometer data as the reference measurement, in detecting increases in days ≥30 minutes of MVPA. Conclusion The SIM is a potentially useful assessment tool for evaluating change in MVPA, particularly when device-based measures or longer self-report measures are not feasible.