I think I believe the USA number, here...
Country,
Other | Total
Cases | New
Cases | Total
Deaths | New
Deaths | Total
Recovered | Active
Cases | Serious,
Critical | Tot Cases/
1M pop | Tot Deaths/
1M pop |
|---|
| USA | 81,943 | +13,732 | 1,177 | +150 | 1,864 | 78,902 | 2,112 | 248 | 4 |
| China | 81,285 | +67 | 3,287 | +6 | 74,051 | 3,947 | 1,235 | 56 | 2 |
| Italy | 80,589 | +6,203 | 8,215 | +712 | 10,361 | 62,013 | 3,612 | 1,333 | 136 |
| Spain | 56,347 | +6,832 | 4,154 | +507 | 7,015 | 45,178 | 3,166 | 1,205 | 89 |
| Germany | 43,646 | +6,323 | 262 | +56 | 5,673 | 37,711 | 23 | 521 | 3 |
| Iran | 29,406 | +2,389 | 2,234 | +157 | 10,457 | 16,715 | 2,746 | 350 | 27 |
| France | 29,155 | +3,922 | 1,696 | +365 | 4,948 | 22,511 | 3,375 | 447 | 26 |
...in the sense that, now that we're testing for it, we can see it.
I would love to see 5 new columns in this table:
- NT => Sev. Sym (New Tests Conducted of Persons With Severe Symptoms)
- NT => Mod. Sym (New Tests Conducted of Persons With Moderate Symptoms)
- NT => Mild Sym (New Tests Conducted of Persons With Mild Symptoms)
- NT => Asym+C (New Tests Conducted of Asymptomatic Persons with a known contact with a person who tested positive)
- NT => Asym+NC (New Tests Conducted of Asymptomatic Persons with a NO known contacts with persons who tested positive)
My guess is that, in the U.S.A., we'd see something like the following numbers (at least, in their rough proportions):
NT => Sev. Sym: 20,000 tests
NT => Mod. Sym: 5,000 tests
NT => Mild Sym: 1,000 tests
NT => Asym+C: 500 tests
NT => Asym+NC: 63 tests
In short, I'm betting that we actually had 500,000 cases last week, of which we'd tested and gotten positives for 65,000 or so, testing mostly only the folk who came in with severe symptoms (and not all of them). And I'm also betting that as more tests become available, we'll be gradually expanding the testing into other categories (first the Moderate Symptom folks, then the Mild Symptoms, and eventually the Asymptomatics). But I bet we haven't done that very much yet. Maybe some in New York; but certainly not nationwide.
Even more briefly: I think the chart's "(Known) New Cases" is as much a consequence of "New Testing," as of "New Infections."
Maybe I'm wrong. I'd love to know, one way or the other.
But without knowing who we were testing last week, and how often,
and who we're testing this week, and how often,
the increasing caseload number seems more potentially deceptive than helpful.
I think I agree with
@creativespiral about focusing on the number of
new deaths.
To focus on deaths seems logical because, for any given
age range, with any given
rough health profile, the lethality of COVID-19 shouldn't vary much from country-to-country provided the country in question has a first-world medical system that isn't yet overburdened.
(By "age range" I mean 60-69, and by "rough health profile," I mean things like "in perfect health" vs. "one chronic condition" vs. "two chronic conditions.")
Now those are all things for which we should be able to obtain reliable case data (excluding places like China or Iran). With that data in-hand, we could do some cool things:
1. We could "normalize" the ages and health profiles from one country to another. (That is, if the virus were 11.006% lethal in Italy, but the median age in Italy is 52, and the median age in the U.S. is 38, then we could break down the Italian sample-set by age and rebalance the ages to estimate its likely lethality in the U.S.)
2. We know the numbers of deaths for most countries pretty reliably; whereas no country (so far as I know) has yet randomly tested the general population to estimate the overall percentage of the population infected. But
if so much as one or two countries would do that, we could then get a ratio of actual infections to deaths. And then we could look at the deaths in any country, and estimate how many infections that represented.
