Where you live may reveal how long you’ll live

For the first time, a new study has estimated life expectancy across neighbourhoods in Canada’s largest cities – the first in Canada to drill down below the levels of provinces and cities to census tracts, the smallest level of geography that can support estimates of life expectancy.

The results are disturbing.

Along with my colleagues, we examined 3,348 census tracts in Canada’s 15 largest cities, as well as ones with robust data in the largest U.S. cities. Variations in life expectancy across geographic areas are widely observed, and are generally associated with socio-economic factors such as income poverty rates and median household incomes. While overall life expectancy in Canada was about three years longer among the cities studied, we found differences as much as 20 years between one census tract and another, all within a single large city.

Of Canada’s 15 largest cities, Victoria had the widest range in life expectancies, while Oshawa had the smallest range. Victoria, St. Catharines, Ont., and Winnipeg had the largest associations between life expectancy and income poverty rates and median family incomes, while Vancouver and Toronto had the weakest associations.

Our results showed substantial differences in life expectancy among cities within the same province. These results are especially notable given the frequent claims that health care is provincial jurisdiction, and hence their responsibility. The implication of these findings is that something about municipalities also has important effects on life expectancy and health inequalities.

Perhaps surprisingly, we also found the range of life expectancies across census tracts in Canada’s largest cities could be just as wide as within the six largest U.S. cities, where variations ranged up to about 20 years.

However, there was a major difference when it came to income poverty rates: the U.S. cities showed a much stronger link between life expectancy and poverty than in Canada’s six largest cities.

This is puzzling. On the one hand, the largest cities in both countries show similarly wide variations in life expectancy across census tracts. On the other, income poverty rates accounted for much more of these health inequalities in the U.S. cities than in Canada.

What could be driving these health inequalities in Canada, if not the usual socio-economic status measures?

While Canada has purportedly “universal” health care (despite major gaps in dentistry, drugs and long-term care) that is certainly much more universal than in the U.S., this cannot be a strong explanation. There were differences in Canadian life expectancy inequalities across cities within the same province.

Another possibility is that Canada is significantly more egalitarian than the U.S. (though less egalitarian than some other OECD countries), which is evident not only in terms of income inequality, but also in scholastic achievement. Further, and notwithstanding the American rhetoric about being the land of opportunity, Canadians experience twice the rate of movement up (and down) the income ladder from one generation to the next than in America. And while there is increasing attention in the U.S. to “deaths of despair,” there is growing evidence that a more pervasive cause of shorter life expectancies is chronic stress.

But something other than national-level differences in inequality must be at work. The significantly stronger associations within each of the six largest U.S. cities between life expectancy and poverty rates could be ascribed to greater racial segregation across U.S. cities. While Canada also suffers from racial and other forms of discrimination, recent analyses suggest something more general: there is more socio-economic segregation across U.S. neighbourhoods, including but not exclusively by race.

In part, these broader differences in socio-economic segregation may derive from very different local government structures. In the U.S., wealthy suburbs can “opt out” of supporting poorer neighbourhoods in the same city by creating their own local governments and school boards. In contrast, recent waves of municipal and school-board amalgamation in Canada’s largest cities mean local public goods such as schools are more equitably distributed, with many fewer locally elected authorities.

The findings in our study raise important questions. While Canada’s largest cities appear to have substantially reduced the link between income poverty and life expectancy compared to their American counterparts, why are the variations within them as wide as those in the U.S.? Even though health care is a provincial and territorial responsibility, why can these patterns be so different from one city to the next within the same province? Is municipal and school-board amalgamation a “hidden gem” of public-health policy?

Now that we have the data, addressing these questions could reframe the discourse on health and health care policy in Canada.

Michael Wolfson is a former assistant chief statistician at Statistics Canada and current member of the University of Ottawa’s Centre for Health Law, Policy and Ethics.