The problem with using absolute risks is that they vary by population, so if you use an absolute risk difference in one paper it's unlikely to apply anywhere else 2/
-
-
Show this thread
-
Say you find that people who take a medication have a 50% decreased risk of heart attack, but your study is in young women who rarely have heart attacks The finding is still really important! 3/
Show this thread -
Your absolute risk might only decrease from 0.001% to 0.0005% per year in that sample, but remember - risk ratios tend to stay relatively constant. If you apply it to 80 year olds with heart disease it might turn into a massive difference 4/pic.twitter.com/INhNQrvIcM
Show this thread -
That being said, the relative risk ratio will be meaningless to >90% of people for whom the yearly risk of a heart attack is low, which makes it a huge problem if it's presented without context in a headline 5/pic.twitter.com/951EZDQ0IZ
Show this thread -
TL:DR - relative risks are a problem in headlines, not in research. They can be very useful as well as very misleading! 6/6
Show this thread
End of conversation
New conversation -
-
-
What are your thoughts when a drug company mixes absolute risk of side effects and relative risk of benefits?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.