Comment by tomerico | Hacker News Reader

Comment by tomerico | Hacker News Reader

The projections are ignoring a very important datapoint - there is significant evidence that COVID-19 does not have a positive R0 in warm climates. This means that in many areas, there is no need to flatten the curve - it will flatten naturally in the next month or two. On the contrary, by aggressively shutting down the spread now, we are leaving ourselves more vulnerable to winter outbreak which will be hard to stop.

Now the really terrible part is that the economic effect of the mass closure is going to grow exponentially. So layoffs start small and grow more massive week by week. And unlike COVID-19, that economic effect will not subside with the warming weather...

Here is some evidence:

Compare the slow growth in cases per day (green line) in CA and FL (warm) vs NY and WA (cold):

https://covid-19.direct/state/CA

https://covid-19.direct/state/FL

https://covid-19.direct/state/NY

https://covid-19.direct/state/WA

See for example how in Malaysia there was an outbreak due to a very large Muslim gathering with foreign travelers, that seems to diminish every day (indicated an R factor < 1) - https://www.worldometers.info/coronavirus/country/malaysia/ The same is true for Qatar.

Same examples all over the world for warm countries: https://www.worldometers.info/coronavirus/

Maybe I don't understand. The number of confirmed cases is going up exponentially in all of the warm locations you mention. This is in spite of a lack of testing. How is that evidence of R < 1?

What we are seeing is that the warmer the place, the slower the growth, with tropical or very warm places (Malaysia, Singapore, Taiwan, India, etc.) showing an R0 < 1

The point is that places like California will experience slow growth which will also decline quickly with the warming weather, not that every place has negative growth right now.

Negative R0 meaning the virus actually heals people? ;)

Maybe you should read this from an actual epidemiologist: https://ccdd.hsph.harvard.edu/will-covid-19-go-away-on-its-o...

more data is needed; the way that article is written is very casual and this is not a peer reviewed paper.

Fair, but the author is about as well-respected an epidemiologist as you will find.

The article is explaining what the default assumption should be. Empirical data might change that view.

Is there any evidence beyond the layperson analysis you are submitting here in this comment?

My critique of this analysis is that there is no way you are controlling for all of the factors: different starting dates for community spread, uneven testing / reporting of cases, random luck due to super-spreading events, differences in behaviors between regions, and so on.

How can you reliably use confirmed cases to suggest a trend given the disparity between states in test availability, population density, and timeline of introduction?

We are looking just at the trend. And the testing methodology that different states take is just a multiplicative constant of the confirmed test. And exponentially growing function will make it negligible.

We are looking just at the trend. And the testing methodology that different states take is just a multiplicative constant of the confirmed cases. And exponentially growing function will make it negligible.

Consider that the evidence you are presenting does not show the actual influence of temperature on the virus itself. I would argue (anecdotally) that warmer climates mean people are less confined in general, therefore adding some element of "natural" social distancing.

Annoyingly, I would think it could proxy details that are reverse correlated. Colder means more gloves and fiddling with more clothes that cover the face, as an example.

Which is to say, I accept it is complicated. If it correlates with something, that is just another question. Not likely an answer in itself.

Brazil is not seeing any slowdown due to high temperatures. Looking at your cities on very different climates, temperature seems to not be a factor at all.

City density, by its turn, is huge.

Reply on news.ycombinator.com