Single versus multi-item self-assessment of sedentary behaviour: A comparison with objectively measured sedentary time in nurses.

Objectives: To compare sedentary time (ST) measured by self-report using a single question from the short-form International Physical Activity Questionnaire (SF-IPAQ), 18-items from the Sedentary Behaviour Questionnaire (SBQ) and objectively using an accelerometer among a large sample of nurses.

Design: Cross-sectional.

Methods: Participants wore an ActiGraph GT3X accelerometer (≥4 days, ≥10h/day) and self-reported usual day sitting using the IPAQ and sitting in different modes using the SBQ. Measures were compared using correlations, a Friedman test with Wilcoxon signed-ranks tests for pairwise comparisons, linear regression and Bland-Altman plots.

Results: A total of 313 nurses (95% female; mean±SD: age=43±12 years) from 14 hospitals participated. Participants self-reported sitting for a median of 240min/day using the SF-IPAQ and 328min/day using the SBQ. Median ST measured by the ActiGraph was 434min/day. All measures were weakly correlated with each other (ρ=0.31-40, ps<0.001). Limits of agreement were wide between all measures. Significant proportional bias between the ActiGraph and the SF-IPAQ and SBQ existed, suggesting that with greater amounts of ST, there is greater disagreement between the self-report and objective measures.

Conclusions: In a sample of nurses, self-reported ST using the SF-IPAQ and SBQ was significantly lower than that measured by accelerometer. A single-item tool performed more poorly than a multi-item questionnaire. Future studies should consider including both objective and self-report measures of ST, and where possible use a tool that quantifies ST across multiple domains, define a 'usual day' and are meaningful for those with daily schedule variations such as among shift-worker populations.