There are different reasons for this. E.g. RCTs are the purview medical statistics more focused on treatment, and not as relevant for epi, which is more closely related to public health and prevention.
-
-
Show this thread
-
There is also a perspective that somehow observational approaches to learning from data are somehow more intellectually challenging, and RCTs are simple and easy. Thus if you can design/analyze a cohort study, you are already qualified to design a RCT and analyze the data.
Show this thread -
So I just want to suggest, from the perspective of a trained epidemiologist who was thrown into the deep end of clinical trials and is only now starting to feel comfortable doggie paddling, it would be a very good idea to make RCTs a foundational component of epi training.
Show this thread -
First, it's embarrassing when one of my fellow epidemiologists assumes they understand RCTs when they have never actually given them more than 2 seconds of thought.
Show this thread -
Second, you might actually want to use an RCT eventually! The entire premise of epi seems to be "let's find the modifiable causes and then intervene" but once you get to the "intervene" part...you will probably want to run a trial if you can.
Show this thread -
And finally, I can't overstate how much having to think more about RCTs has helped me refine my views on epidemiological questions/designs/analyses. Different contexts, but sharing the same fundamentals for causal and statistical inference, leads to informative contrasts.
Show this thread
End of conversation
New conversation -
-
-
You should get better Epi training then?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
The practice of conducting an RCT is indeed on the disputed border or Epidemiology, but the theory is not and most good courses should teach this. You can't understand many of the challenges and pitfalls with observational data without reference to an RCT.
-
I think a lot more can and should be done bringing a formal causal inference perspective to the conduct and analysis of RCTs.
@EpiEllie's work in particular shows the importance of that. But the RCT 'community' is quite hostile to people on the 'outside' dabbling... - Show replies
New conversation -
-
-
There is this myth that any clinician can run an RCT. One still needs a team to do so. Epi has a planning role. Too many non-epi dab into epi work. At end of study clinicians start looking for 'who is available' to look at the data. My 1st Q now is 'has ICH-GCP been followed?'
-
This Tweet is unavailable.
- Show replies
New conversation -
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.