The last four World Health Organization Coronavirus disease situation reports, 63-66, show an almost flat number of new tested-and-reported COVID-19 cases worldwide each day:
- 46,484 (report 67; March 27; bold face because added as an update)
- 62,514 (report 68; March 28)
- 63,159 (report 69; March 29)
- 58,411 (report 70; March 30)
- 57,610 (report 71; March 31)
- 72,736 (report 72; April 1)
- 72,839 (report 73; April 2)
- 75,853 (report 74; April 3)
- 79,332 (report 75; April 4)
- 82,061 (report 76; April 5)
- 77,200 (report 77; April 6)
- 68,766 (report 78; April 7)
- 73,639 (report 79; April 8)
- 82,837 (report 80; April 9)
- 85,054 (report 81; April 10)
- 89,657 (report 82; April 11)
- 85,679 (report 83; April 12)
- 76,498 (report 84; April 13)
- 71,779 (report 85; April 14)
- 70,082 (report 86; April 15)
- 76,647 (report 87; April 16)
- 82,967 (report 88; April 17)
- 85,678 (report 89; April 18)
- 81,153 (report 90; April 19)
- 72,846 (report 91; April 20)
- 83,007 (report 92; April 21)
- 73,920 (report 93; April 22)
- 73,657 (report 94; April 23)
- 81,529 (report 95; April 24)
- 93,715 (report 96; April 25)
- 84,900 (report 97; April 26)
- 85,530 (report 98; April 27)
- 76,026 (report 99; April 28)
- 66,276 (report 100; April 29)
- 71,839 (report 101; April 30)
- 84,771 (report 102; May 1)
- 91,977 (report 103; Saturday, May 2)
- 82,763 (report 104; May 3)
- 86,108 (report 105; May 4)
- 81,454 (report 106; May 5)
- 71,463 (report 107; May 6)
- 83,465 (report 108; May 7)
- 87,729 (report 109; May 8)
- 95,866 (report 110; Saturday, May 9)
- 61,563 (report 111; May 10)
- 88,891 (report 112; May 11)
- 82,591 (report 113; May 12)
- 81,577 (report 114; May 13)
- 100,284 (report 123; May 22)
Original text, before the data points in bold were added: Almost, but not quite flat, right? After four days it grew from around 40,000 to around 50,000. But consider these data against the background of a rapidly ramping up testing infrastructure. More people are being tested, not just celebrities and political elites with mild symptoms and peasant hospital inpatients. If dramatically more people are being tested every few days (and therefore fewer mild cases never get noticed/recorded as COVID-19), don’t the above data suggest that the actual number of new cases (humans actually infected as opposed to the proper subset of humans tested and reported as infected) is going down?
- March 25 report: 679 new cases
- March 26 report: 579 new cases
- March 27 report: 823 new cases (5,750 tests)
- March 28 report: 1,107 new cases (5,678 tests)
- March 29 report: 698 new cases (4,017 tests)
- March 30 report: 797 new cases (3,727 tests)
- March 31 report: 868 new cases (4,142 tests)
- April 1 report: 1,118 cases (4,803 tests)
- April 2 report: 1,228 cases (4,870 tests)
- April 3 report: 1,436 cases (6,354 tests)
- April 4 report: 1,334 cases (5,838 tests)
- April 5 report: 764 cases (3,137 tests, a Sunday)
- April 6 report: 1,337 cases (4,492 tests, lab staff back to work on a Monday?)
- April 7 report: 1,365 cases (4,915 tests)
- April 8 report: 1,588 new cases (6,167 tests)
- April 9 report: 2,151 new cases (7,447 tests)
- April 10 report: 2,033 new cases (7,414 tests)
- April 11 report: 1,886 new cases (6,404 tests; 87 deaths)
- Sunday, April 12 report: 2,615 new cases (7,954 tests; 70 deaths)
- April 13 report: 1,392 new cases (5,319 tests; 88 deaths)
- April 14 report: 1,296 new cases (4,502 tests; 113 deaths)
- April 15 report: 1,755 new cases (5,472 tests; 151 deaths)
- April 16 report: 2,263 new cases (8,750 tests; 137 deaths)
- April 17 report: 2,221 new cases (7,971 tests; 159 deaths)
- Saturday, April 18 report: 1,970 new cases (8,062 tests; 156 deaths)
- Sunday, April 19 report: 1,705 new cases (5,435 tests; 146 deaths)
- April 20 report: (switch to new format): 1,566 new cases (7,157 tests; 103 deaths)
- April 21 report: 1,556 new cases (5,974 tests ; 152 deaths)
- April 22 report: 1,745 new cases (5,090 tests; 221 deaths)
- April 23 report: 3,079 new cases (14,614 tests; 178 deaths)
- April 24 report: 4,956 new cases (20,127 tests (backlog resolved); 196 deaths)
- April 25 report: 2,379 new cases (11,632 tests; 174 deaths)
- April 26 report: 1,590 new cases (9,255 tests; 169 deaths)
- April 27 report: 1,524 new cases (8,787 tests; 104 deaths)
- April 28 report: 1,840 new cases (9,613 tests; 150 deaths)
- April 29 report: 1,963 new cases (11,118 tests; 252 deaths)
- April 30 report: 1,940 new cases (10,029 tests; 157 deaths)
- May 1 report: 2,106 new cases (13,989 tests; 154 deaths)
- Saturday, May 2 report: 1,952 (9,358 tests; 130 deaths)
- … see archives
- May 13 report: 1,165 (8,536 tests; 174 deaths)
- May 22 report: 805 new cases (10,158 tests; 80 deaths)
3/27 analysis: Not an obvious exponential growth process and, even if it were, patients aren’t generally tested until admitted to a hospital (i.e., current growth in cases reflects a growth in infections that happened 1-2 weeks ago)
[Update, evening 3/27, text message from friend: “I asked [physician wife] how her nurse friend is doing at the hospital. She said she has not worked in a week because the ER is so dead she can’t get hours. The hospitals are empty.”]
3/28 analysis: The Massachusetts data are worrisome. The number of reported tests stayed constant from March 27-28, but the percentage of positives grew. We have no information about when samples were taken, however, so this growth could have occurred several days earlier. (Test results show up in this report on the day that the tests are completed.) The WHO data suggest that the answer to the question posted in the headline is “no” (though it is tough to say given that the WHO situation report does not say how many tests are being performed worldwide).
3/29 analysis: Massachusetts testing actually fell. Only 17 percent of tests were positive, about the same as the average of the two previous days. Linear growth is a better fit than exponential growth. The WHO data suggest weak exponential growth.
3/30 analysis: Massachusetts testing fell again, but 21 percent tested positive. Suggests that doctors are now better at figuring out who needs to be hospitalized (since it is hospitalization that leads to being tested). WHO data show a slightly reduced number of both cases and deaths.
3/31 comment: Where’s the exponential growth?
4/1 comment: Okay, maybe we do have (slow) exponential growth in both worldwide and Massachusetts cases.
4/2 comment: The world is doing better than Massachusetts, but if Farr’s law is in operation, it is tough to guess the top of the bell curve.
4/3 comment: Sweden (open) has 2X the population of Denmark (locked down), so the 519:279 ratio of new cases works out to roughly the same as the population. The ratio of deaths (43:19) is also roughly the same as population ratio. So it looks as though the Swedish epidemiologist’s prediction that government policy would have a minimal effect is proving correct. For the world overall, it has been 11 days and still the number of new cases has not doubled. WHO is still not publishing data regarding how many tests are being performed, so there is no way to know if this slow doubling corresponds to a 4X increase in testing and a falling number of actual new infections or a steady rate of testing and a doubling-every-two-weeks rate of actual new infections. Massachusetts data suggest doubling every 8-9 days. They also confirm the wisdom of the New Yorkers who fled via Gulfstream to their Nantucket mansions: only 9 cases on the island.
4/4 comment: Sweden continues to have roughly the same number of new cases, per capita, as Norway and Denmark, and Swedish hospital/ICU capacity remains sufficient (and public). It seems to be essentially impossible to die from COVID-19 in Norway. Do they have a better treatment that they’re not sharing? Or do they record the death of someone with underlying conditions differently? Or are they keeping people on ventilators beyond the point at which the Swedes would have given up hope?
4/5 comment: Sweden has 365 new cases; Denmark has 320. Keeping in mind that Sweden has 2X the population, this suggests that the Danish “lockdown” is not “making a difference” as rich Americans like to say. Massachusetts is back to the same number of new cases as 9 days ago. Farr’s law, but with an extremely poor fit to a Gaussian?
Monday, 4/6 comment: Sweden has 387 new cases; Denmark has 292. I.e., Sweden continues to have a lower per-capita new-case rate despite the lack of a lockdown. Sweden even has a lower per-capita death rate, 28:18. Spain and the UK are contributing hugely to the world total of new cases, each with roughly 6,000. Massachusetts numbers are back where we’d expect them to be, perhaps due to reduced lab activity on Sunday.
4/7 comment: Sweden has 376 new cases; Denmark has 312. The per capita new death ratio is no longer favorable to Sweden, however, at 4.75X (small sample, however; Denmark had only 8 deaths). Globally, COVID-19 has killed more than 72,000 people, i.e., about the same as the number of Americans killed each year by taxpayer-funded (via Medicaid) opioid addiction and overdose. But the new case count seems to be declining in accordance with Farr’s law. The Massachusetts numbers are worrisome. New cases are stubbornly high and 356 people have died so far (roughly the same, per capita as in Sweden; i.e., a poor argument for our state’s school closure and other shutdown efforts).
4/8 comment: Sweden has 487 new cases; Denmark has 390 (i.e., more per capita in the locked-down country). Worldwide new cases are still on a plateau. Massachusetts cases continue to grow, consistent with a doom-and-gloom forecast from University of Washington that Massachusetts will end up worse than New York, adjusted for population. Fully 2.6 percent of people in Massachusetts with confirmed cases are already dead, suggesting that we are not better at caring for patients than were the doctors working in Wuhan (death rate roughly 1.4 percent). As in China, though, our data are skewed by limited testing.
4/9 comment: Sweden is at 726 new cases; Denmark 331 (comparable when adjusted for population size). Worldwide new cases continue on a plateau. Massachusetts in shutdown seems to delivering the exponential growth that the media was hoping to see from wide open Sweden.
4/10 comment: Sweden has 722 new cases; Denmark 233. Sweden has more new cases adjusted for population, but Massachusetts, with more than 2,000 new cases, is a dramatically higher rate (roughly 4X). Deaths in Massachusetts, 96, represent a higher rate, adjusted for population, than Sweden’s, at 106.
4/11 comment: Let-it-burn Sweden has 544 new cases; lock-it-down Denmark has 184. Sweden is 1.5X the rate, adjusted for population. Sweden had 77 deaths versus 87 in shutdown Massachusetts (i.e., MA residents are dying at 1.6X the rate when adjusted for population).
4/12 comment: Sweden has 466 new cases; locked-down Denmark has 177. Sweden suffered 17 deaths; Denmark 13. I.e., adjusted for its 2X population, wide-open Sweden has a higher rate of new cases and a lower rate of deaths. Massachusetts has gone off the rails compared to Sweden. With 2,615 new cases, Massachusetts has 8X the new case rate of Sweden’s. With 70 deaths, locked-down Massachusetts has 6X the death rate of Sweden.
4/13 comment: Sweden at 332/12; Denmark at 178/13. Massachusetts at 1392/88.
4/14 comment: Sweden at 465/20; Denmark at 144/12; Massachusetts at 1296/113. Adjusted for population, Massachusetts has 4X the new cases and 8X the deaths.
4/15 comment: Sweden at 497/114; Denmark at 193/14; Massachusetts at 1,755/151. It seems that there was a substantial lag in reporting deaths in Sweden due partly to a three-day Easter holiday weekend. Even if all of the 114 deaths had occurred in the preceding 24 hours, the rate of new deaths in Massachusetts is substantially higher when adjusted for population. Total deaths in Massachusetts are at 1,108 versus 1,033 in Sweden, a 1.5X higher rate in the shut-down region of Massachusetts versus the up-and-running get-it-over-with region of Sweden.
4/16 comment: Sweden at 482/170; Denmark at 170/10; Massachusetts at 2,263/137. Total deaths in Sweden are 1,203 versus 309 in Denmark (2X the rate, adjusted for population). With 1,245 total deaths, shut-down Massachusetts still has a higher rate than mostly-open Sweden and a higher rate of new deaths.
4/17 comment: Sweden at 613/130; Denmark at 321/12; Massachusetts at 2,221/159. It is getting tougher to argue that the Danish shutdown was ineffective (but maybe Denmark will suffer its infections starting in a few weeks; kids in Denmark went back to school on Wednesday). And it is getting tougher to argue that the Massachusetts shutdown was effective (but maybe we won’t have a second wave since we were so incompetent at slowing down the first wave?). Sweden has had 1,333 total deaths without shutting down. Massachusetts shut down and has had 1,404 deaths (1.5X the rate).
4/18 comment: Sweden 676/67; Denmark 194/15; Massachusetts 1,970/156.
4/19 comment: Sweden 606/111; Denmark 193/12; Massachusetts 1,705/146. Every article that I saw today in the nationwide media about a COVID-19 victim featured a healthy person cut down in his/her/zer/their 30s or 40s. Out of 146 Massachusetts deaths total, just 4 were under age 60 and all had preexisting conditions. The median age of a person who died was “80s”.
Monday, 4/20 comment: Sweden 563/29; Denmark 142/9; Massachusetts new format 1,566/103. As nobody seems to die in Sweden on a Sunday or holiday, the 29 number should be ignored for now. With 1,809 total deaths, Massachusetts continues to have a higher (1.67X) rate than Sweden (1,540). The chart of “currently hospitalized” has been more or less flat since April 14. So it may be time to reduce our coronapanic level regarding finding hospital services if we need them (but we can coronapanic about something else!).
Thursday, 4/23 comment: Sweden 682/172; Denmark 217/14; Massachusetts 3,079/178. The Massachusetts death rate to date is now 1.75X that of Sweden (2,360 deaths versus 1,937). The number of hospitalized patients continues to be more or less level. The number of deaths per day is also more or less level:
(it seems as though there is a substantial delay from death to report). There was a huge jump in testing as of today’s report, which means the number of newly reported cases is not comparable to the old numbers.
Friday, 4/24 comment: Sweden 751/84; Denmark 161/10; Massachusetts 4,946(!)/196. The death rate in Massachusetts is now 1.8X the Swedish rate. The MA report explains that the high new case count “reflects a correction of a reporting error made by Quest, a national commercial laboratory, which affected case counts in multiple states including Massachusetts” and that these cases actually go back to April 13. The average age of a person who died while testing positive for COVID-19 is 82. Out of 2,556 deaths, only 1 has been of a person under age 30.
Wednesday, 4/29 comment: Sweden 695/81; Denmark 153/7; Massachusetts 1,963/252. With 3,405 deaths so far in Massachusetts versus 2,355 in Sweden, the Massachusetts death rate, halfway through our second month of lockdown, is 2X Sweden’s. India has experienced only 31,332 cases and 1,007 deaths, despite the vast population and impracticality of a true lockdown.
With roughly the same number of “new cases” worldwide as at the beginning of the month, despite what must be an increased testing capacity, I am prepared to guess that the number of new cases is actually declining. Deaths in today’s report were 5,376. Deaths reported on April 1 were 4,193. (If my guesses continue to prove wrong, this will qualify me for a job as an epidemiologist at the University of Washington’s Bill Gates-funded Institute for Health Metrics and Evaluation!)
Saturday, May 2 comment: Sweden 428 new cases/67 deaths; Denmark 153/8; Massachusetts 1,952/130. The number of people in the hospital in Massachusetts has been trending down for five days, consistent with news from a neighbor who is an ER nurse at a community hospital (“things are quieting down”). We’re back to where things were circa April 14. The worldwide new case count of 91,977 sounds high, but it is also roughly the same as where it was on April 11.
Wednesday, May 13 comment: Sweden 602 new cases/57 deaths; Denmark 78/-6 (six Danes who had died from Covid-19 actually rose from the dead); Massachusetts 1,165/174. In Massachusetts to date, 5,315 people have been tagged as having died from/with Covid-19 (average age 82 with underlying conditions holding steady at 98.4%). In Sweden, it is 3,313, which means the Massachusetts death rate is 2.3X Sweden’s.
Wednesday, May 22 comment: Sweden 649 new cases/40 deaths; Denmark 65/7; Massachusetts 805/80. Deaths to date in Sweden: 3,871; in Massachusetts: 6,228. The Massachusetts death rate is still 2.3X Sweden’s. The epidemic seems to be winding down here in Massachusetts, proof that the virus will burn itself out no matter how incompetent a group of people are? Looking county-by-country, it doesn’t seem as though the curve shapes are radically different in any of the plague lands:
How about just Massachusetts? From NBC:
Speaking of states, could it be that New York is the safety place in the U.S. to be right now? Via the magic of exposing almost everyone to the virus back in March, the state looks like it is near the end of an almost textbook epidemic curve:
Do we think that Governor Cuomo heroically saved those still alive in NYC or that the virus simply ran out of suitable hosts?