3 ms·
This would be bad, but effect sizes are important. If it imposes a 1/1000 incremental odds of developing diabetes, that would be terrible, but not as bad as its
by asdfasgasdgasdg 6y ago
This would be bad, but effect sizes are important. If it imposes a 1/1000 incremental odds of developing diabetes, that would be terrible, but not as bad as its 1/100 incremental odds of death.
- khuey 6y agoIf the deaths come to people who are elderly or otherwise close to death anyways and the diabetes comes to a bunch of otherwise healthy 18 year olds the impacts could be different than mere effect sizes would indicate. (Type 1 diabetes is like -10 years to life expectancy).