Article ID: | iaor2016274 |
Volume: | 35 |
Issue: | 11 |
Start Page Number: | 2069 |
End Page Number: | 2086 |
Publication Date: | Nov 2015 |
Journal: | Risk Analysis |
Authors: | Lordan Grace, Pakrashi Debayan |
Keywords: | risk, statistics: empirical, medicine, behaviour |
In Britain, it is recommended that, to stay healthy, adults should do 150 minutes of moderate‐intensity physical activity every week. The recommendations provided by the U.K. government, however, remain silent in regard to the type of activity that should be done. Using the annual Health Survey for England we compare how different types of physical activities predict a person's weight. In particular, we consider clinically measured body mass index and waist circumference. We document mean slopes emanating from ordinary least squares regressions with these measures as the dependent variables. We show that individuals who walk at a brisk or fast pace are more likely to have a lower weight when compared to individuals doing other activities. Additionally, we highlight that the association between physical activity and weight is stronger for females and individuals over the age of 50. Our overall conclusions are robust to a number of specifications.