Pandemic Update March 28

2020-03-29Posted by xaminmo

CV19 Pandemic Update for 2020-03-28 (Saturday Night).

I’m thinking that the increase in testing that we have done may simply be what’s required to keep up with our infection spread rate, and maybe does not actually dilute our results. The US mortality rate actually just started to go up again.

The numbers are close to estimates, but non-PRC is 6k low, Texas is 2% high, and Maine is 1.5% high. Either the rest of the world is slowing down, with the future estimates coming back in line, or this was an anomaly which will self-correct soon, with an over-sized day.

2020-03-28 Saturday
1-Day Projection: WW 666020/30928; PRC 82012/3301; NON 584008/27627; IT 92840/10156; US 123266/2067; TX 2400/32; ME 182/1
1-Day Distancing: WW 666020/30928; PRC 82012/3301; NON 584008/27627; IT 92840/10156; US 123859/2000; TX 2390/33; ME 182/1
Actual Numbers:   WW 660706/30652; PRC 81999/3299; NON 578707/27353; IT 92472/10023; US 121478/2026; TX 2455/30; ME 211/1

I only include the numbers for PRC because it fits my chart. I am fairy cetrain those numbers are bogus after seeing that their funeral homes appear to be cremating 4-8 times the normal number of bodies. That matches the supposition that we only get 14% of their numbers.

ERR US Testing:   POS 107,568 NEG 561,770 PEND 59,784 HOSPITAL 23,397 DEAD 2,585
1-Day US Testing: POS 107,345 NEG 569,288 PEND 64,889 HOSPITAL 19,587 DEAD 2,013
Actual Testing:   POS 118,234 NEG 617,470 PEND 65,712 HOSPITAL 16,729 DEAD 1,965

I had a formula error and pasted the wrong line for the estimate. Not too far off from the updated estimate, but the real numbers are much further off than I expected.

2020-03-29 Sunday Estimate
1-Day Projection: WW 736981/34604; PRC 82101/3302; NON 654880/31302; IT 98859/10999; US 145164/2596; TX 3112/35; ME 265/1
1-Day Distancing: WW 736981/34604; PRC 82101/3302; NON 654880/31302; IT 98859/10999; US 144607/2494; TX 2934/37; ME 265/1
1-Day US Testing: POS 126,269  NEG 659,435  PEND 70,178  HOSP 23,397  DEAD 2,585

2020-04-01 7-day Projection April 1
7-Day Projection: WW 1,169,645/61,852; PRC 82,134/3,338; NON 1,087,512/58,514; IT 157,313/19,560; US 477,863/6,398; TX 9,131/162
UPDATE            WW 1,097,956/54,118; PRC 82,357/3,321; NON 1,015,599/50,796; IT 132,717/16,052; US 333,315/5,929; TX 7,263/90; ME 468/5

The worldwide numbers are lower by 70k. The US 477k is from a one-day 175% several days ago, and it has not rolled out of the average. If I dilute that, the estimate is 300k Note that the US is not tracked as part of the world here.

7-Day Distancing: WW 1,169,645/61,852; PRC 82,134/3,338; NON 1,087,512/58,514; IT 157,313/19,560; US 232,880/4,603; TX 5,875/72
UPDATE            WW 1,097,956/54,118; PRC 82,357/3,321; NON 1,015,599/50,796; IT 132,717/16,052; US 237,600/4,513; TX 5,003/68

The distancing algo still tracks close, with 1.5% difference.

7-Day US Testing: POS 134,035 NEG 728,160 PEND 100,028 HOSP 68,262 DEAD 4,770 Overrun: 04-06 335k
UPDATE            POS 150,376 NEG 785,328 PEND  83,576 HOSP 64,012 DEAD 5,886 Overrun: 04-06 342k

The US testing chart looks like hospital overrun may still be April 6 (32% of US hospital beds).

Texas projections are the same except the 1.2m point moves out 1 day.

04-07 29k TEXAS ICU OVERRUN STAGE 1
04-06 36k TEXAS VENT OVERRUN STAGE 1
04-13 146k TEXAS ICU STG3
04-14 178k TEXAS VENT STG3 & Hosp STG1
04-17 400k may be the inflection point for Texas if 14% like China, but our death rate is lower than theirs or the world.
04-18 532k TEXAS HOSP STG3
04-22 1.2m Inflection point for R0 of 2.2 is 60% with detection rate of 39.08

The distancing formula update shows TX ICU one day earlier, and hospital stage 3 one day later.

04-10 2929 TEXAS ICU OVERRUN STAGE 1
04-12 1787 TEXAS VENT OVERRUN STAGE 1
04-20 2929 TEXAS ICU OVERRUN STAGE 3
04-21 1787 TEXAS VENT OVERRUN STAGE 2
04-21 26633 TEXAS HOSPITAL OVERRUN STAGE 1
04-25 400k may be the inflection point for Texas if 14% like China, but our death rate is lower than theirs or the world.
04-27 4550 TEXAS VENT OVERRUN STAGE 3
04-28 26633 TEXAS HOSPITAL OVERRUN STAGE 2
05-02 47441 TEXAS HOSPITAL OVERRUN STAGE 3
05-02 1.2m Inflection point for R0 of 2.2 is 60% with detection rate of 39.08

Here are the milestones from 04-23 data for comparison to track our progress flattening the curve:

04-06 33k TEXAS ICU OVERRUN STG 1
04-07 45k TEXAS VENT OVERRUN STG 1
04-10 150k TEXAS VENT/ICU STG 2 & Hosp STG 1
04-13 366k TEXAS VENT/ICU STG 3 & Hosp STG 2
04-15 660k TEXAS Hosp STG 3

Milestones are based on guesstimates: TX 28.7m pop; 2.9 beds per 1000 in TX (83230); 32% unoccupied (26633) / 11% are ICU (2929); 6.7% ventilators (1787); 10% limited function ventilation (2663) / 5-15% conf need hosp(532k-177k @ overrun); 2-10% ICU (29k-146k); 1-5% need vent (36k-178k)

Testing has ramped up substantially, every 2 days another 20k per day. We are at 110k per day as of 2020-03-28 16:00 EDT. We are expecting another 50k/day from Abbott Labs. That probably is just going to help keep up with demand, and not dilute deaths by confirmations. US mortality rates are up again, at 1.67%,

Our mortality rate is still about a third of the rest of the world, with TX going down to 122%, and the US going up again today to 167% (155% yesterday, 144% prior).

Inflection point for outbreaks with an R0 of 2.2 is 60% of the population, not 60% of the confirmed cases. If we are catching a third of the cases, then uncontrolled, exponential climb would stop when confirmed cases were 20% of the population. If we are catching a 6th, then 10%.

Social distancing data lags by 3 days. Statistics lag by 1 day. Detection lags by 9-12 days. Impact delay of SD on Confirmations MAY BE 9 days, or it may be longer. The correlation is hard to suss out, but it may be that it will become more evident as we get more days of data.

Spreadsheet is updated, and downloadable here:
https://drive.google.com/file/d/1vocCN445AZyVBBLsv0kJR8ZDP9DM0UST/view

Projection Spreadsheet

2020-03-22Posted by xaminmo

Google link is https://drive.google.com/file/d/1vocCN445AZyVBBLsv0kJR8ZDP9DM0UST/view View/comment only. I don’t know what broke yet, or if everything works right.

0317 Projection Update

2020-03-17Posted by xaminmo

Estimate for today was within 1% again, except Italy was 2% lower on both infections and deaths. That is two days of increasingly lower rates than projected. GOOD JOB ITALY! Their March 22 numbers are on track to be less than the original 1week projection. Death rates are still high, at almost 8%. It’s hard to bring that down, especially since their population is 9 years older than the US, China, South Korea, etc.

The US has been consistently higher than projections. Authorities won’t test for COV19 if you test positive for influenza first. It’s likely the US reporting rate is less than the 14% that China had. On March 22, the US will be higher than the original 1week projection. Death rates are at 1.68%, which is suspiciously low. I am concerned that perhaps deaths are being attributed to other things, or testing is being avoided so that the stats look better. It could also be a new strain of the virus, but I would think that would get a lot of promotion from the government.

I’m having a discrepancy between World vs PRC + nonPRC, with WORLD going high. This may be a random error, but I’m wondering if China’s +30/day is false information. They have kicked out foreign press, and are in a propaganda war with the White House over who’s at fault. The difference is bigger than I would expect could come from the US’s reporting anomalies.

Because of reporting deficiencies, the real number infected is 6-8x the formal numbers, and that typically equates to about a week into the future as a SWAG for real but unreported cases. If you are still going out of the house with non-family members for recreational purposes, you are being selfish, and are a menace to society. If you are doing so to provide essential services, then please be extra careful.

Some of you may have noticed that, if the detection rate is only 14%, then that means the mortality rate is also a sixth of what we think. Yes, but it still spreads too fast, and the absolute numbers of people needing hospital care to survive still would overwhelm hospitals, leading to much higher death rates. So, even if the mortality rate is a flat 0.5% for everyone under age 64, and hospitalization needs are still only 2%, we still need to flatten the curve. At the current rate, and assuming 14% reporting rate, the US is on track to have the whole population infected in one month, with 6 million needing hospital care, and about a million needing external ventillation. We have about a million beds, and about 100k ventillators, with another 60k of old and less effective types. Also, if everyone is sick, who can be the caregivers?

Projections for tomorrow are WW 214086/8769; 81083/3243 PRC; 134076/5591 Non-PRC; 35476/2903 IT; 8901/137 US; 142/1 TX. I think we can expect the US to be 1-2% above projection, and Italy 1-2% below projection. Texas should have additional deaths any day now, but both TX and the US are still tracking substantially lower death rates than expected. I’m hoping to see a slow-down for the US next week. It seems a lot of people got the message for today, and have been reducing their out-of-house socializing.