Speaking of miscarriages and fetal demise and analysis I did with Dr. Thorp, I noticed where CDC coders entrusted to extract a "symptom" from the narrative, incorrectly transcribed/assigned a spontaneous abortion to the report instead of foetal death, or even premature baby death!
Write-ups would say, "I had a miscarriage at 23 weeks" or "I had a miscarriage at 5 months". Both of these statements would be incorrect statements of miscarriage as defined (<19.999 weeks). Since I never found a report where I could say this was a good coder with the wisdom to know the difference , I started to think how much of this "coding" was being "automated" and how much was needed to be seen by human eyes? Surely they are automating some of the low hanging fruit easy stuff and do not need a coder to code everything? Especially the report that are simple administration errors only as an example.
Excellent write-up Aravind, it's worthy of me doing a write-up of explanation of this exact topic (data mining and diagnosing) causing the rift between Liz&Jess and myself. The full back story is here: https://www.vaersaware.com/post/openvaers-com-the-myocarditis-stats-the-full-story.
Speaking of miscarriages and fetal demise and analysis I did with Dr. Thorp, I noticed where CDC coders entrusted to extract a "symptom" from the narrative, incorrectly transcribed/assigned a spontaneous abortion to the report instead of foetal death, or even premature baby death!
Write-ups would say, "I had a miscarriage at 23 weeks" or "I had a miscarriage at 5 months". Both of these statements would be incorrect statements of miscarriage as defined (<19.999 weeks). Since I never found a report where I could say this was a good coder with the wisdom to know the difference , I started to think how much of this "coding" was being "automated" and how much was needed to be seen by human eyes? Surely they are automating some of the low hanging fruit easy stuff and do not need a coder to code everything? Especially the report that are simple administration errors only as an example.
I would like the Florida Grand Jury to demand access to all the data and employ some thinkers to look at the clear damage signals.
This is out of place but on your question elsewhere about lots like SDCH8, please see the 'tame' file in https://hawkvaers.com/download/2022_09_22_vaers_lot_cleaning_v.997gh/output/, sorting by 'cleaned' perhaps.