A lot of people in the world of evidence appraisal use the phrase "Garbage-In Garbage-Out" (GIGO) to describe bad meta-analyses I thought I'd briefly explain what this means 1/n
-
Show this thread
-
2/n This all comes back to what a meta-analysis *does* In essence, it's very simple. Think of your bog-standard arithmetic mean (average) - a bunch of numbers added together then divided by the totalpic.twitter.com/K4R0jJKbzO
1 reply 7 retweets 42 likesShow this thread -
3/n This is the most basic form of meta-analysis - lump all the numbers together into an average. Simple, but obviously flawedpic.twitter.com/tLQTpXSdvz
1 reply 6 retweets 33 likesShow this thread -
4/n Let's say you have a study of 10,000 people with a value of 5 for your outcome of interest, and another study with a value of 10 people with a value of 15 The crude average is (5+15)/2 = 10, but that's clearly not useful
1 reply 6 retweets 34 likesShow this thread -
5/n So what we do instead is weight the studies This is our meta-analytic model - a *weighted* average that gives more weight to bigger/better studies
1 reply 6 retweets 39 likesShow this thread -
6/n A common method of doing this is to use the inverse of the variance of the studies. This essentially uses the confidence intervals of each study as a weighting tool, with tighter intervals getting a higher weightpic.twitter.com/yW7WF1YY8o
1 reply 8 retweets 43 likesShow this thread -
7/n Here's an example from the ivermectin literature - this is simply a weighted average where the weighting is derived from the inverse of the variance (calculated from the confidence intervals)pic.twitter.com/ux75gk6HFz
1 reply 7 retweets 36 likesShow this thread -
8/n But you can immediately see the issue here - this weighting is DERIVED FROM THE DATA If the data is unreliable, the weighting is meaningless!pic.twitter.com/7Noojnz1iQ
2 replies 8 retweets 71 likesShow this thread -
That's a very complex question! Depends on the study
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.