[finolex1]

Conventional life expectancy measures are calculated based on the death rates at each age of the residents in that place. Therefore the death rates of people at 80+ or 90+ years may be affected by their lifestyles 30-50 years ago, so it's not always an accurate measure of how long you would expect a child just born to survive in that place.

AFAIK it's not weighted by the time people live in that place. Perhaps areas with lots of affluent retirees will have distorted life expectancy rates that may not be reflective of people who are born and spend their lives in that area.

[dainichi]

Thanks for that explanation. So if a county has, say, a lot of hospices (and people change their residency when they go there, not sure how that works in the US), that will affect the expectancy negatively, saying nothing about how living in that county affects longevity.