![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
On the news they sometimes mention that the "peak" in cases is expected to happen around a certain date, usually within a few weeks. Along with a stated or hinted-at indication that after then, things can begin going back to normal, ie. social restrictions eased.
https://www.wcnc.com/article/news/health/coronavirus/heres-when-covid-19-could-peak-in-the-carolinas/275-62ed6956-892e-4585-9434-73936bf728ac
https://www.msn.com/en-us/health/medical/here-s-when-covid-19-could-peak-in-oregon/ar-BB122H7O
I've wondered how they come up with those dates; what they base it on, as it seems quite optimistic to me, and doesn't mesh with what I've internalized from my reading.
But today I may have figured it out... What follows is a big generalization, as there are many more factors involved than what I'm considering here, and the numbers I'm using are all very rough estimates. And maybe my thinking is all wrong. But at least it makes sense to me at the moment.
I had read that without social distancing, the number of positive cases doubles about every 2 days. That corresponds to an increase of about 1.4 per day.
With social distancing, the rate of increase here in SC has gone down to under 1.1 per day. That is based on a spreadsheet I've been keeping, with daily stats taken from SC's DHEC COVID-19 Data page. (It's a loose estimate, as I'm not even sure if I should compare the total number of cases each day or only the number new cases each day.)
When the rate of infection goes under 1, then there will no longer be an increase, and that is when we will be past the "peak". (or maybe 2 weeks after that time, based on incubation times).
That can also be thought of as the time when each infected person infects fewer than one other person, on average.
The current number of confirmed positive cases in SC is 2792, but the estimate of actual cases is 18000, or about 6 to 7x as many.
SC's population is about 5 million.
The peak for SC is estimated as April 24, which is 14 days from now.
Given a current increase of about 1.1 per day, if we raise 1.1 to the 14th power, that would give an increase of 3.8 in those 14 days.
Multiplying 18000 current cases by 3.8 gives about 68000....
Actually, using these numbers, it doesn't add up at all.
But when I was thinking about it earlier today, I estimated the peak being in 30 days.
1.1 to the 30th power is 17.
I estimated there being at least 10 actual cases per diagnosed case. So I estimated about 25000 current cases.
25000 * 17 would give us about 425,000 positive cases in 30 days.
425,000 would be close to 10% of the population.
If 10% of the population has had the disease and can no longer catch or spread it, that should reduce infection rates by about 10%, bringing the current 1.1 rate of increase down to below 1.
And that's why that date would be the peak, even though only 10% of the population has caught the virus by then, and 90% are still vulnerable to it.
So it has to rely on social distancing being maintained. Otherwise the infection rates could go back up above 1 again, just like it was in the beginning.
Based on the FAQ on this page: http://www.healthdata.org/covid/faqs,
the people who come up with those peak date estimates may not even be considering the same factors as I am. It sounds like their models are based on when the peak was reached in other locations, and based on the amount of social distancing in each place.
But the FAQ does also state:
So that confirms that things can't simply go back to normal after we've passed the (first) peak.
Updated to add, 2020/04/10:
Maybe the infection rate will go down below 1, even simply due to the social distancing and hand-washing, etc. measures. Maybe it simply hasn't been in place long enough to see that yet. My city gave a stay-at-home order about 12 days ago but the state-wide orders came only 9 (some closures) and 4 days ago (further closures). That could explain the peak coming earlier even with smaller numbers of people infected.
https://www.wcnc.com/article/news/health/coronavirus/heres-when-covid-19-could-peak-in-the-carolinas/275-62ed6956-892e-4585-9434-73936bf728ac
https://www.msn.com/en-us/health/medical/here-s-when-covid-19-could-peak-in-oregon/ar-BB122H7O
I've wondered how they come up with those dates; what they base it on, as it seems quite optimistic to me, and doesn't mesh with what I've internalized from my reading.
But today I may have figured it out... What follows is a big generalization, as there are many more factors involved than what I'm considering here, and the numbers I'm using are all very rough estimates. And maybe my thinking is all wrong. But at least it makes sense to me at the moment.
I had read that without social distancing, the number of positive cases doubles about every 2 days. That corresponds to an increase of about 1.4 per day.
With social distancing, the rate of increase here in SC has gone down to under 1.1 per day. That is based on a spreadsheet I've been keeping, with daily stats taken from SC's DHEC COVID-19 Data page. (It's a loose estimate, as I'm not even sure if I should compare the total number of cases each day or only the number new cases each day.)
When the rate of infection goes under 1, then there will no longer be an increase, and that is when we will be past the "peak". (or maybe 2 weeks after that time, based on incubation times).
That can also be thought of as the time when each infected person infects fewer than one other person, on average.
The current number of confirmed positive cases in SC is 2792, but the estimate of actual cases is 18000, or about 6 to 7x as many.
SC's population is about 5 million.
The peak for SC is estimated as April 24, which is 14 days from now.
Given a current increase of about 1.1 per day, if we raise 1.1 to the 14th power, that would give an increase of 3.8 in those 14 days.
Multiplying 18000 current cases by 3.8 gives about 68000....
Actually, using these numbers, it doesn't add up at all.
But when I was thinking about it earlier today, I estimated the peak being in 30 days.
1.1 to the 30th power is 17.
I estimated there being at least 10 actual cases per diagnosed case. So I estimated about 25000 current cases.
25000 * 17 would give us about 425,000 positive cases in 30 days.
425,000 would be close to 10% of the population.
If 10% of the population has had the disease and can no longer catch or spread it, that should reduce infection rates by about 10%, bringing the current 1.1 rate of increase down to below 1.
And that's why that date would be the peak, even though only 10% of the population has caught the virus by then, and 90% are still vulnerable to it.
So it has to rely on social distancing being maintained. Otherwise the infection rates could go back up above 1 again, just like it was in the beginning.
Based on the FAQ on this page: http://www.healthdata.org/covid/faqs,
the people who come up with those peak date estimates may not even be considering the same factors as I am. It sounds like their models are based on when the peak was reached in other locations, and based on the amount of social distancing in each place.
But the FAQ does also state:
"By the end of the first wave of the epidemic, a substantial proportion of the population of the United States and EEA countries are likely to still be susceptible to the disease and thus measures to avoid a second wave of the pandemic prior to vaccine availability will be necessary."
So that confirms that things can't simply go back to normal after we've passed the (first) peak.
Updated to add, 2020/04/10:
Maybe the infection rate will go down below 1, even simply due to the social distancing and hand-washing, etc. measures. Maybe it simply hasn't been in place long enough to see that yet. My city gave a stay-at-home order about 12 days ago but the state-wide orders came only 9 (some closures) and 4 days ago (further closures). That could explain the peak coming earlier even with smaller numbers of people infected.