Atomic wrote: ↑Tue Sep 01, 2020 10:18 am
One thing that bothers me about the "Number of Cases" type reporting is the (lack of) definition for a "Case".
Agreed - it would be better if the published reports were specific. The CDC does have some recommendations, but they acknowledge that different states and counties are reporting based on their own locally-chosen criteria.
- Tested positive? (asymptomatic but part of the body count)
Historically, I believe this is probably the most common definition of a "case" for communicable diseases. The person in question is infected, may or may not be "sick" (symptomatic) and (depending on the disease) may be contagious.
By this standard, Typhoid Mary was a "case" of typhoid fever. She was never symptomatic (as far as we know) but she was contagious. She refused to accept that she was contagious because she felt fine, kept working as a cook, and was responsible for infecting a large number of others.
- Sick / exhibiting symptoms? (most likely definition)
The problem with this, is that it misses two categories of people who are infected (and often contagious): those who are pre-symptomatic (but will begin exhibiting symptoms in the next few days), and those who are asymptomatic and will remain so.
- Hospitalized? (unduly restrictive count)
The better reports distinguish these. The ones I've seen come out of my county (Santa Clara Co., in California) distinguish cases, hospitalizations, hospitalizations requiring intensive care (an ICU bed), and deaths.
It's bad enough that graphs and numbers are being flung about liberally, but no analysis or interpretations are being offered.
You often have to drill down past the "for quick public consumption" press reports, and look one level deeper into the actual data reported by the states and counties.
"Highest Number Of Cases Reported Yet!" - Report released summarizing the past few months of test results = big one day number. Bleah.
I'd agree with "bleah" if it's just a statistical fluctuation. If it's part of an increasing trend in the area in question, it's an "uh oh" rather than a "bleah".
As to what constitutes a "case" for reporting purposes, here's some information from the CDC's current guidelines document (
https://www.cdc.gov/nchs/data/icd/COVID ... -final.pdf). Emphasis-in-bold added by me.
1) COVID-19 Infections (Infections due to SARS-CoV-2)
a) Code only confirmed cases
Code only a confirmed diagnosis of the 2019 novel coronavirus disease (COVID-19) as documented by the provider, documentation of a positive COVID-19 test result, or a presumptive positive COVID-19 test result. For a confirmed diagnosis, assign code U07.1, COVID-19. This is an exception to the hospital inpatient guideline Section II, H. In this context, “confirmation” does not require documentation of the type of test performed; the provider’s documentation that the individual has COVID-19 is sufficient.
Presumptive positive COVID-19 test results should be coded as confirmed. A presumptive positive test result means an individual has tested positive for the virus at a local or state level, but it has not yet been confirmed by the Centers for Disease Control and Prevention (CDC). CDC confirmation of local and state tests for COVID-19 is no longer required.
If the provider documents "suspected," "possible," "probable," or “inconclusive” COVID19, do not assign code U07.1. Assign a code(s) explaining the reason for encounter (such as fever) or Z20.828, Contact with and (suspected) exposure to other viral communicable diseases.
To sum that up - the CDC says to use the reporting code for COVID-19 if there's a positive test result, or if the provider (the doctor or hospital) declares a confirmed diagnosis of the disease. The latter can, i presume, be done on the basis of signs-and-symptoms and medical/social history, or by exclusion (that is, if they have cough and fever and pneumonia but test negative for flu and bacteria, and were in contact with a known COVID patient).
Considering that testing for COVID-19 is still quite spotty (shortage of materials and backlogs in the lab), it seems quite reasonable to me to accept a doctor's diagnosis of COVID-19 even in the absence of a formal test result. If you can't test a sick person because you don't have a test available, that doesn't somehow make the illness "not COVID". It'd be nice if it did, but that's sloppy "magical thinking" - don't test, and the "cases" go away. The real world doesn't work that way.