It is not. Large numbers have no bearing whatsoever on residual confounding
-
-
Replying to @GidMK @EdoajoEric
You're wrong and you know it. The residual confounding have to be enormous to explain the difference, so big that it must be evident and it isn't. Same in Peru, same in Uttar Pradesh India, same in other places. Residual confoundings do explain Together trial result indeed.
2 replies 0 retweets 5 likes -
Replying to @jjchamie @EdoajoEric
You appear to genuinely have no idea what confounding is. This is a reasonable primer if you're interested in knowing why it's an issue for this analysis but not the Together trial: https://sph.unc.edu/wp-content/uploads/sites/112/2015/07/nciph_ERIC11.pdf …
1 reply 0 retweets 5 likes -
Replying to @GidMK @EdoajoEric
Certainly there are residual confounders, but not large enough to even moderately change the results. If so, chances are they will change the results for the better.
1 reply 0 retweets 3 likes -
Replying to @jjchamie @EdoajoEric
That's simply inaccurate - the residual confounding is likely to impact the results. For example, who is more likely to get tested and then get a pack? Etc
1 reply 0 retweets 5 likes -
-
Replying to @jjchamie @EdoajoEric
Nope - epi 101, often socio-economics drive choices like these. Entirely possible that the richer people were both more likely to get a test AND receive a pack. The point is not that this DID happen, but that it could change the results if it had
2 replies 0 retweets 5 likes -
Replying to @GidMK @EdoajoEric
You're making assumptions ignoring third world reality. Rich people in Mexico are a small group, they don't go to the public heaths system. They go to private doctors and buy their own drugs.
1 reply 0 retweets 3 likes -
Replying to @jjchamie @EdoajoEric
Rich is contextual, and in this context is shorthand for "more socio-economically advanced than average"
2 replies 0 retweets 5 likes -
Replying to @GidMK @EdoajoEric
You can't explain this 76% improvement with this shaky factors. Plus, the factors can play in both directions. The subsequent improving at national level after rolling out test and treat with ivermectin is another evidence. The CFR improved 5 fold.
1 reply 1 retweet 4 likes
CFR is largely based on testing numbers, so very unreliable. You can often explain even large differences such as 76% with residual confounding what you're saying is incorrect
-
-
Replying to @GidMK @EdoajoEric
CFR is 80% less, and testing is not the explanation. Testing is very low. Test positivity rate over 40%. CFR is unreliable in the UK indeed. Mass testing school kids twice per week lower the numbers.
1 reply 0 retweets 1 like -
Replying to @jjchamie @EdoajoEric
In Mexico, daily tests appear to have roughly doubled over the time period of the study. That would usually result in a lower CFR even if there was no effect
2 replies 0 retweets 3 likes - 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.