Background The Health Survey for England (HSE) includes the extensive Physical Activity and Sedentary Behaviour Assessment Questionnaire (PASBAQ) in some years of data collection to monitor physical activity (PA) levels in England. Due to funding and interview length constraints, this cannot be included each year; policy-makers are keen to consider a shorter PA questionnaire for the annual survey. We assessed the validity of summary measures designed to rank participants according to their levels of PA and sedentary behaviour using the short-version of the International Physical Activity Questionnaire (IPAQ), compared with the PASBAQ.
Methods 1252 participants aged 16+ completed both instruments (PASBAQ main-interview, IPAQ nurse-visit) in the fourth quarter of HSE 2012. IPAQ-assessed time spent sitting on weekdays, walking, and walking + moderate-to-vigorous PA (MVPA) were compared with PASBAQ-assessed sedentary time, walking, and MVPA. Durations in each activity were summarised using the median due to skewed distributions. Boxplots showed IPAQ-assessed MVPA according to PASBAQ-assessed adherence to current MVPA recommendations (≥150 min/week). Continuous variables were grouped into tertiles; the weighted Kappa statistic (κ) was used to estimate concordance between the questionnaires. Analyses were stratified by sex, age-group (16–64, 65+), and time between interview and nurse-visit (<2, 2–4, >4 weeks).
Results Differences in each activity by subgroup were similar for both instruments. MVPA time was higher using IPAQ versus PASBAQ (median 895 and 360 min/week respectively among men; 525 and 215 min/week among women). Time spent walking was similar across both instruments, except for higher levels among participants aged 65+ using IPAQ (210 and 105 min/week). IPAQ-assessed sitting time appeared lower than PASBAQ among all groups.
Of those participants in the highest tertile of PASBAQ-assessed MVPA, 55% of men and 59% of women were classed in the highest tertile using IPAQ-assessed MVPA. Equivalent figures for the lowest tertile were 55% and 48%. Boxplots showed increases in IPAQ-assessed MVPA time according to PASBAQ-assessed levels of adherence to current recommendations. Kappa coefficients showed moderate concordance among all groups and activities, ranging from 0.26 to 0.40.
Discussion The IPAQ and PASBAQ had different reference periods (7 day versus 28 day), with an average gap of 4½ weeks between data-collection. Despite considerable disagreement, IPAQ-assessed MVPA increased with increasing levels of PASBAQ-assessed MVPA. Future work will examine concordance between the instruments in more detail, including assessments of the total volume of activity.
- physical activity validation
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.