Open Access Highly Accessed Research

Predictors of change differ for moderate and vigorous intensity physical activity and for weekdays and weekends: a longitudinal analysis

Kirsten Corder12*, Christopher Craggs12, Andrew P Jones13, Ulf Ekelund24, Simon J Griffin12 and Esther MF van Sluijs12

Author Affiliations

1 UKCRC Centre for Diet and Activity Research (CEDAR), Institute of Public Health, University of Cambridge, Box 296, Forvie Site, Robinson Way , Cambridge CB2 0SR, UK

2 MRC Epidemiology Unit, Institute of Metabolic Science, Box 285, Addenbrooke’s Hospital, Hills Road, Cambridge CB2 0QQ, UK

3 Norwich Medical School, University of East Anglia, Norwich NR4 7JT, UK

4 Department of Sports Medicine, Norwegian School of Sports Sciences, PO Box 4014, Ullevål Stadion, Oslo 0806, Norway

For all author emails, please log on.

International Journal of Behavioral Nutrition and Physical Activity 2013, 10:69  doi:10.1186/1479-5868-10-69

Published: 28 May 2013



Predictors of physical activity (PA) change are rarely investigated separately for different PA intensities and for weekdays/weekends. We investigated whether individual-level predictors of one-year change in objectively-measured physical activity differ for moderate PA (MPA) and vigorous PA (VPA) and for weekends and weekdays.


Accelerometer-assessed PA (mins) was obtained at baseline and +1 year (n = 875, 41.5% male, Mean ± SD baseline age: 9.8 ± 0.4 years-old). Potential predictors (n = 38) were assessed at baseline from psychological (e.g. self-efficacy), socio-cultural (e.g. parent support) and environmental domains (e.g. land use). Associations between predictors and change in MPA (2000–3999 counts/minute (cpm)) and VPA (≥4000 cpm) separately for weekdays and weekends were studied using multi-level linear regression. Analyses were adjusted for school clustering, sex and baseline PA.


Weekend PA declined (MPA decline 4.6 ± 21.8 mins/day; VPA decline: 2.1 ± 20.1 mins/day; both p < 0.001) whereas weekday PA did not significantly change. Higher baseline PA and being a girl were associated with greater PA declines in all four outcomes; remaining predictors differed for MPA and VPA and/or weekdays and weekends. Family logistic support was associated with less of a decline in weekend MPA (CI 95%) 0.15 (0.05, 0.25) and VPA 0.19 (0.09, 0.29), and peer support with less of a decline in weekday MPA 0.18 (0.02, 0.34) and VPA 0.22 (0.06, 0.38).


Results highlight the relevance of investigating predictors of PA change separately for different PA intensities and for weekdays/weekends. In addition to continued focus on school PA promotion, more effort to target interventions during weekends, such as in the family and community appears important. Encouraging peer support to increase weekday PA and targeting parent support for weekend PA may be health promotion priorities.

Longitudinal; Behaviour change; Childhood; Determinants; Physical activity; Predictors