Re: Covfefe-19 part 6: Waiting For Advice from Kid Rock and Tiger Woods now.
Is there still a lag on test results? Should we look at yesterday's number testing positive and think of it as people tested a week ago?
This is a fascinating and difficult question of relational data.
Take a table of patients. A patient's attributes are unique identifier, dob, gender, resident address, work address, etc. Patients are linked to each other many-to-many through genetic kinship, residential propinquity, work propinquity, friendship, etc.
The next table is tests. A test's attributes are unique identifier, test type, date administered, analyzing lab, result.
There are testers, labs, diagnosticians, etc.
The next table is diagnoses, with unique id, patient, date, diagnosis, diagnostician.
All the data is important because we may see patterns which can direct us to conclusions about medical realities or sources of error. In wolverine's question what we want is an aggregated report of tests by date, with result. To expose heterogeneous granularity we may want to break down tests by the patient's observed symptoms -- perhaps productive cough is not dispositive but hypoxia has high lethality. We should also be tracking treatments (ventilation and interval, vaccine id, administration date, strength, etc) and their results.
Hopefully this is all being done by hospitals on the micro level, the CDC nationally and WHO globally, as there is an entire science and infrastructure of medical data analysis, and it's enough in the weeds that it is probably not vulnerable to direct political manipulation by bad actors.