I wonder if anyone ever did a mortality rate study on the close friend or confidantes of these folks?
any actuarians out there to take a statistical stab at this?