> anything other than differences in levels of exercise/activity I've seen claims that having few friends shows up in mortality worse than than a pack-a-day sm

> anything other than differences in levels of exercise/activity I've seen claims that having few friends shows up in mortality worse than than a pack-a-day smoking habit. Of course that's all hand-wavy statistical controls, but then again so is the stuff on exercise and calories.