Original Article

International Journal of Obesity (2009) 33, 1427–1436; doi:10.1038/ijo.2009.174; published online 6 October 2009

Joint associations of physical activity and sedentary behaviors with body mass index: results from a time use survey of US adults

G F Dunton1, D Berrigan2, R Ballard-Barbash2, B Graubard3 and A A Atienza1

  1. 1Health Promotion Research Branch, Behavioral Research Program, Division of Cancer Control and Population Sciences, National Cancer Institute, Bethesda, MD, USA
  2. 2Office of the Associate Director, Applied Research Program, Division of Cancer Control and Population Sciences, National Cancer Institute, Bethesda, MD, USA
  3. 3Biostatistics Branch, Epidemiology and Biostatistics Program, Division of Cancer Epidemiology and Biostatistics, National Cancer Institute, Bethesda, MD, USA

Correspondence: Dr GF Dunton, Department of Preventive Medicine/IPR, University of Southern California, 1000 S, Fremont Avenue, Unit 8, Building 5, Room 5229, Alhambra, CA 91803, USA. E-mail: dunton@usc.edu

Received 25 January 2009; Revised 15 June 2009; Accepted 15 July 2009; Published online 6 October 2009.





Obesity risk is negatively associated with physical activity and positively associated with time spent in sedentary behaviors. Yet, it is not known how different combinations of sedentary and active behavior are associated with body mass index (BMI). This study examined the interaction between time spent in physical activity and sedentary behavior on BMI in US adults.



Cross-sectional, data from the 2006 American Time Use Survey.



10984 non-underweight adults (aged 21 + years).



A phone interview assessed all activities performed in the past 24h, height, weight, health status, and other sociodemographic characteristics. Time spent in (1) moderate-to-vigorous leisure-time physical activity (MVPA), (2) active transportation (walking, biking), (3) sedentary leisure activities (TV/movie watching, computer use, playing games, reading), and (4) sedentary transportation (motorized vehicles) was determined from activity coding. BMI was calculated.



After adjusting for age, gender, education level, race/ethnicity, and health status, sample-weighted linear regressions found significant interactions for leisure MVPA × TV/movies, leisure MVPA × playing games, active transportation × sedentary transportation, and active transportation × reading (Ps<0.0001). For example, the group of adults watching <60min per day of TV/movies and engaging in greater than or equal to60min per day of leisure MVPA had lower average BMI compared to the group watching <60min per day of TV/movies and reporting <60min per day of leisure MVPA (P<0.0001). In contrast, for adults watching greater than or equal to189min per day of TV/movies, there was not a significant difference in BMI by time spent in leisure MVPA.



Data from a US time use survey indicate that the strength of the association between certain types of sedentary behavior and BMI varies according to time spent in certain types of physical activity and vice versa.


physical activity, transportation, sedentary behaviors, television watching, body mass index, adults

Extra navigation