The capability to produce a daily count over a large nation is a proof of the functioning existence of a large, distributed infrastructural system which moves around test samples and reagents, coordinates clinical staff + remote techs, and has unimpeded internal information flow.
-
Show this thread
-
When the count goes down, sometimes that means the thing people think it means. Other times it means that the orderly functioning of the supply chain that produces the count has been impaired.
3 replies 9 retweets 58 likesShow this thread -
It is very important that we not mistake Scenario #2 for Scenario #1.
2 replies 0 retweets 32 likesShow this thread -
In Scenario #2, you will see a decrease in the number of tests produced per day, an increase in reported latency between tests being ordered and tests results being ready, and an increase in test positive rate as capacity rationed for patients with known worse clinical outcomes.
2 replies 1 retweet 25 likesShow this thread -
“Explain that rationing.” Suppose that getting access to testing takes a clinician 15 minutes one week but 30 the next. Their rational response is to not conduct discretionary bubble tests (“Probably a waste of time anyhow.”) but still conduct tests for severely ill patients.
1 reply 0 retweets 3 likesShow this thread -
From a provision of care perspective, the person with bilateral pneumonia and severe difficulty breathing needs clinical effort more than someone who came in with a high fever. From an information theory perspective, a test on first patient gives you less new info than second.
1 reply 1 retweet 16 likesShow this thread
“It seems unlikely the doctor will make that decision. Don’t they cover this in...” One of Japan’s foremost epidemiologists was complaining earlier today about how few doctors understand Bayes’ Rule. How often has this been critical during their career, etc etc.
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.