Another commuting study:

I analyze cross-sectional data from the American Time Use Survey (2003-2008) to quantify decreases in health-related activity participation due to commuting and labor time. I examine the data for associations between commute length and time spent in exercising, food preparation, eating, and sleeping behaviors. I augment the data with activity strenuousness scores to test whether physically-draining commutes induce lower-intensity activity substitutions. I find small but highly significant associations consistent with the conjecture that commuting time cost impacts health-promoting behaviors. Each minute spent commuting is associated with a 0.0257 minute exercise time reduction, a 0.0387 minute food preparation time reduction, and a 0.2205 minute sleep time reduction. I find trade-offs due to commuting often exceed trade-offs due to labor time on a per-minute basis. Longer commutes are also associated with an increased likelihood of non-grocery food purchases and substitution into lower intensity exercise activities.

http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1490117