One of the biggest issues I see people make with evidence appraisal is the idea that you have to have a yes/no answer That's not how evidence works It's ALMOST ALWAYS shades of grey
-
-
When epidemiologists review studies, they usually do what's called "evidence grading" This is a formal assessment of the potential biases that the study has, and a judgement on how you should interpret the conclusions
Show this thread -
You might see someone say that the study has a very high risk of bias This means that there are many places in the study where outside influences may have crept in and caused the conclusions to be flawed
Show this thread -
Similarly, a very low risk of bias means that the research was done impeccably and the conclusions are likely robust
Show this thread -
BUT NEITHER OF THESE GIVES YOU A YES OR NO ANSWER
Show this thread -
They are simply judgments on how likely it is that the conclusions are correct. But it's a gradation rather than a dichotomy
Show this thread -
The fact is, most studies are flawed in some way. Even the very best may not be easily usable for many reasons But a potentially biased study can still provide invaluable information
Show this thread -
Some examples of what biased studies have shown us: - smoking causes cancer - seatbelts save lives - asbestos causes mesothelioma - iodine deficiencies cause issues
Show this thread
End of conversation
New conversation -
-
-
Unless you consider the problem of forking paths and file drawers
-
Oh certainly, I'm mostly talking about criticisms of individual studies rather than the meta-criticisms of research as a whole. Industry funding can be a serious issue in terms of what we study and publish!
- 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.