The linear distance calculation between respondents and the list of hospitals is processed by using the Distance function from Django’s gis library. The Distance function takes the coordinates of each respondent’s address and the addresses of hospitals and returns distances that take into account the curvature of Earth. Researchers then select the hospital with the shortest linear distance to the respondent and use the Google Maps API to estimate driving time and driving distance and better reflect local travel patterns.

It looks like the methodology didn't ask respondents, "When you need to go to the emergency room at a hospital, where would you go?" and "How long would it take for you to drive there yourself?" The study took a respondent's address, the address of the closest hospital by straight line distance and then tried to figure out the drive time from there. Good thing insurance companies don't have "out of network" contingencies in their policies.