There are many different levels of association. You point to the example of the number of dietitians increasing as obesity rates increase, so let's talk about that
-
-
Show this thread
-
This is what's known as a simple correlation. I can think of several simple explanations for this fact: 1. dietitians cause obesity 2. more obese people = more demand for dietitians 3. external confounder (i.e. population growth)
Show this thread -
Other examples of simple correlations include that ice-cream is associated with drownings, and that the age of Miss America is associated with murders by steam, hot vapours, and hot objects in the US
Show this thread -
Now, this is VERY DIFFERENT from what the dietary guidelines mean when they say that there is "an association" between adherence to the guidelines and reduced morbidity/mortalitypic.twitter.com/e6msl2lGq5
Show this thread -
In this case, it is basically impossible - for obvious ethical and practical reasons - to run large RCTs on adherence to dietary guidelines This means that is not possible to generate conclusive causal evidence
Show this thread -
What that means is that THE BEST EVIDENCE POSSIBLE is Grade C, or from observational trials In this case, 5 large, well-controlled analyses that generate significant evidence for a point Very different to a simple correlation
Show this thread -
So, here, "an association between" means "very likely causally linked, but it is impossible to entirely eliminate the possibility of confounding"
Show this thread -
While it is true to say that these results are not definitively causal - hence the cautioning statement - it is extremely likely that guidelines improve health This is the opposite of our previous example, which as I identified was almost certainly down other factors
Show this thread -
This is the challenge with interpreting epidemiological evidence, that often you have to infer causation from evidence that can never entirely eliminate the possibility that it is a simple correlation
Show this thread
End of conversation
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.