Of course it doesn't explain everything. The points that you bring up are valid. However, its rational to assume that the distribution of body weight should follow a Gaussian curve, so you can make some assumptions about it. Since obesity is a much larger factor than simply being overweight, then it should account for the bulk of the problem. And since the numbers that I gave you show that the decrease in life expectancy due to obesity should be relatively small, then being overweight or obese can't account completely for the difference in life expectancies. I don't beleive that the differences in lifestyles between Canadians and Americans can account for that difference. A deficiency in American healthcare is more likely the cause, escpecially since infant mortality correlates in the same direction.
In the end, you can always try to come up with anectodal evidence, or objections which are not supported by any data to explain the gap. But infant neglect, and a minor difference in obesity are not beleivable or sufficient enough to cover that difference.
Share this thread with friends: