Working on this paper reminded me of previous work on the U.S., where we also documented urban mortality penalties in 1999 turning into large urban advantages by 2019. It is interesting to see this process unfolding similarly in two very different contexts.
onlinelibrary.wiley.com/doi/full/10....
Interestingly, we did not find a large contribution of circulatory and metabolic diseases, even though they account for more than 1/3 of all deaths. While urban mortality declines for these causes were faster in middle and older adulthood, rural declines were more rapid in young adulthood.
Behind the national trends, we also show large regional differences. For example, the urban penalty was always largest and persisted even in 2023 in the North, but was almost absent in the Southeast since 2010 for males and from before 2006 for females.
Which causes drove the reversal of the urban penalty? For females, declines in respiratory diseases and neoplasms accounted for 66%, with external deaths contributing 21%. For males, external deaths alone contributed ~50%, with neoplasms and respiratory diseases jointly contributing 28%.
This is remarkable because the two approaches rely on very different identifying assumptions, providing sort of a "doubly robust" estimation approach. We hope the methods in this paper could be applied to improve cause-of-death classification for other misreported causes of death.