Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

There are ~1 mil of these kinds of surgeries annually a 1% increase is 10,000 infections that could be prevented by just telling patients not to smoke for a bit. This should be touted as a huge win. Correlation and effect size are independent. You can have extremely strong correlation but a small effect size.


Yeah, except there were only 22,000 cannabis users so it's actually 220 people, not 10,000.


That’s still good data and an awful lot of pain, suffering, time and money.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: