Ewing 2005 - "Can the Physical Environment Determine Physical Activity Levels?"

Ewing, Reid
"Can the Physical Environment Determine Physical Activity Levels?"
Exercise and Sport Sciences Reviews
April 2005; v.33, n.2; pp.69-75
On the Web (pdf)
Relevance: medium high

Ewing seems to be primarily reviewing and re-presenting earlier work, but does so in a concise, easy-to-grasp way. Although he doesn't show any of the data or analysis, Ewing offers elasticity estimates for the effect of the physical environment on physical activity and its mediators. The effects are greatest for the mode share of walking and public transportation. CAVEAT: Ewing shows no statistical tests or analysis, so the accuracy of these estimates is unknown.

Here, an elasticity of X.Y says that for every 1% increase in the sprawl index (indicating a 1% increase in compactness), the variable under consideration also increases by X.Y%.

The walk-mode share increased by 0.93% for every 1% increase in the sprawl index (in compactness)
The transit-mode share increased by 1.78% for every 1% increase in the sprawl index (in compactness)
There also seems to be a threshold effect for New York and Jersey City, whereby at some level of compactness walk- and transit-mode shares increase dramatically more than the models predict.

The percentage of trips made by foot increased by 0.45% for every 1% increase in measures of density or design.

Sprawl appears to have no effect on leisure-time physical activity, as opposed for example, to physical activity gained by walking for errands.

For every 1% increase in the sprawl index, BMI decreased by 0.013%. (That is every 1% increase in compactness decreased BMI by 0.013%.) The author may or may not have controlled for leisure-time activity.