For context: kb posted a WaPo article a few days ago in Chat regarding death counts. I mentioned something about how in my state's reporting system, in order to say that COVID-19 contributes to death, we need a death certificate. Now I feel the need to point out we still list death in the investigation, but there's a specific question as to whether the condition contributed to death. We can't answer that question without the death certificate. That went into talks about how underreporting deaths is just shooting oneself in the foot due to a lack of true counts which will affect federal funding, which then devolved into a slight rant about disease investigation and how much of a shitshow it can be. Chatski doesn't allow paragraphs, but I paragraphed below to make all your lives easier. c_hawkthorne[429121142]: @am_Unition [42935723]: Oh it gets even more fun. Welcome to the wonderful world of Disease Investigation. Alright, so every reportable disease has a case definition. For the most part, they're on the CDC's website. It's buried deep, but the state system links to it and I'm sure other state systems do as well. That website is here: https://wwwn.cdc.gov/nndss/conditions/ Okay, so now that you've looked through that, you'll see COVID-19 isn't even on there! Fortunately, we have our friends over at The Council for State and Territorial Epidemiologists (CSTE) to thank, because they've released a defintion that we are using. We being my area, others within the state and other states, who knows. Again, I can only speak for what we are doing. So CSTE's definition is available in this ten page PDF document: https://www.med.ohio.gov/Portals/0/CSTE%20COVID-19%20Case%20Definition_1.pdf It boils down to -- Confirmed is PCR test through an EUA lab. Probable is other types of tests, including antibody, again from an approved lab, but that person also has to be exhibiting specific symptoms or have an epidemiologic link. Then Probable can also be the same set of specific symptoms and an epi link. The last probable is death certificate (remember that one? Ha). So now we get to what is actually happening, especially with respect to asymptomatic people. We see a massive facility. 150 people. They decide to test everyone. Whoops, they don't do their research and use a lab under EUA review, but not with EUA approval yet. FDA has even said if you're a lab in good standing, whatever that means, and if you validate your test in some magic way, again super vague, that you're fine and don't need anything more. But this specific lab is not EUA approved. As such, right away no case from the facility can be "confirmed" per the definition. Now due to the way we're dealing with healthcare facilities, you live or work in one, you have epi link. So that's not a concern. The problem is symptoms at this point. Every case with a positive result from a test under EUA review who is asymptomatic or not showing the correct symptoms gets counted as "Not a case". We don't even waste time putting it in the system. If that test gets EUA approval, we will not go back and change anything, because we aren't putting them in the system and we're told not to put them in the system at a later date. Now one of my facilities decided to test the entire population. 60-75% of the reports I've recieved so far don't meed clinical defintion, and as such aren't even a case. This way, state gets to say look, we have cases, but we're keeping case counts in facilities low. There's a massive ordeal around long term care facilites and COVID. Being able to keep case counts low but not too low is a good thing. It feels like that's what they're trying to do.
The FDA is kind of weird in this way. Well, I guess it makes sense on their end to not be overly prescriptive with how to accomplish something. I bet b_b can speak to this too from his perspective, I'm only speaking to interactions I've had with the FDA within the Food & Beverage industry... They really don't tell you how to do something unless you bug them about it long enough, and figure out how to bug the right person. This is both good and bad - there's more than one way to do most things, after all, and all that needs to be provided is validation of the effectiveness of the work performed to meet the FDA regulation and guidance. It allows flexibility in manufacturing and in some cases, actual innovation! The downside is...a lot of the time it leaves it up what the fuck are you actually asking me to do!?!? and a lack of trust in the system...FDA has even said if you're a lab in good standing, whatever that means, and if you validate your test in some magic way, again super vague, that you're fine and don't need anything more. But this specific lab is not EUA approved.
Yeah my regular job is regulatory. I just tell people they're wrong and they have to fix stuff, but I can't tell them how to fix it, that's up to them. Otherwise we might be playing too hard into their finances which becomes a hairy situation fast. We do conflict of interest paperwork annually and even the year after we leave as well. I don't envy the FDA these days. They're stuck between a rock and a hard place. I'm not saying it's wrong to not say how something has to be done, but they can be less vague about it. That lack of trust... the more I learn about everything the more I feel it.