PreprintPDF Available

The first three months of the COVID-19 epidemic: Epidemiological evidence for two separate strains of SARS-CoV-2 viruses spreading and implications for prevention strategies

Authors:

Abstract and Figures

About one month after the COVID-19 epidemic peaked in Mainland China and SARS-CoV-2 migrated to Europe and then the U.S., the epidemiological data begin to provide important insights into the risks associated with the disease and the effectiveness of intervention strategies such as travel restrictions and lockdowns (“social distancing”). Respiratory diseases, including the 2003 SARS epidemic, remain only about two months in any given population, although peak incidence and lethality can vary. The epidemiological data suggest that at least two strains of the 2020 SARS-CoV-2 virus have evolved during its migration from Mainland China to Europe. South Korea, Iran, Italy, and Italy’s neighbors were hit by the more dangerous “SKII” variant. While the epidemic in continental Asia is about to end, and in Europe about to level off, the more recent epidemic in the younger US population is still increasing, albeit not exponentially anymore. The peak level will likely depend on which of the strains has entered the U.S. first. The same models that help us to understand the epidemic also help us to choose prevention strategies. Containment of high-risk people, like the elderly, and reducing disease severity, either by vaccination or by early treatment of complications, is the best strategy against a respiratory virus disease. Lockdowns can be effective during the month following the peak incidence in infections, when the exponential increase of cases ends. Earlier containment of low-risk people merely prolongs the time the virus needs to circulate until the incidence is high enough to initiate “herd immunity”. Later containment is not helpful, unless to prevent a rebound if containment started too early. About the Author Dr. Wittkowski received his PhD in computer science from the University of Stuttgart and his ScD (Habilitation) in Medical Biometry from the Eberhard-Karls-University Tübingen, both Germany. He worked for 15 years with Klaus Dietz, a leading epidemiologist who coined the term “reproduction number”, on the Epidemiology of HIV before heading for 20 years the Department of Biostatistics, Epidemiology, and Research Design at The Rockefeller University, New York. Dr. Wittkowski is currently the CEO of ASDERA LLC, a company discovering novel interventions against complex (incl. coronavirus) diseases from data of genome-wide association studies.
Content may be subject to copyright.
m
The first three months of the COVID-19 epidemic:
Epidemiological evidence for two separate strains of SARS-
CoV-2 viruses spreading and implications for prevention
strategies
Two epidemics of COVID-19
KNUT M. WITTKOWSKI1*
1 ASERA LLC, New York, NY
* Corresponding author:
E-mail: knut@asdera.com (KMW)
Abstract
About one month after the COVID-19 epidemic peaked in Mainland China and SARS-CoV-2
migrated to Europe and then the U.S., the epidemiological data begin to provide important insights
into the risks associated with the disease and the effectiveness of intervention strategies such as
travel restrictions and social distancing. Respiratory diseases, including the 2003 SARS epidemic,
remain only about two months in any given population, although peak incidence and lethality can
vary. The epidemiological data suggest that at least two strains of the 2020 SARS-CoV-2 virus
have evolved during its migration from Mainland China to Europe. South Korea, Iran, Italy, and
Italy’s neighbors were hit by the more dangerous “SKII” variant. While the epidemic in continental
Asia is about to end, and in Europe about to level off, the more recent epidemic in the younger
US population is still increasing, albeit not exponentially anymore. The peak level will likely
depend on which of the strains has entered the U.S. first. The same models that help us to
understand the epidemic also help us to choose prevention strategies. Containment of high-risk
people, like the elderly, and reducing disease severity, either by vaccination or by early treatment
of complications, is the best strategy against a respiratory virus disease. Social distancing or
“lockdowns” can be effective during the month following the peak incidence in infections, when
the exponential increase of cases ends. Earlier containment of low-risk people merely prolongs
the time the virus needs to circulate until the incidence is high enough to initiate “herd immunity”.
Later containment is not helpful, unless to prevent a rebound if containment started too early.
About the Author
Dr. Wittkowski received his PhD in computer science from the University of Stuttgart and his ScD
(Habilitation) in Medical Biometry from the Eberhard-Karls-University Tübingen, both Germany.
He worked for 15 years with Klaus Dietz, a leading epidemiologist who coined the term
“reproduction number”, on the Epidemiology of HIV before heading for 20 years the Department
of Biostatistics, Epidemiology, and Research Design at The Rockefeller University, New York. Dr.
Wittkowski is currently the CEO of ASDERA LLC, a company discovering novel treatments for
complex diseases from data of genome-wide association studies.
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -2-
Table of Content
Introduction 3
Materials and Methods 4
Data 4
Methods and models 4
Statistical and Bioinformatics Methods ................................................................ 4
Epidemiological Models ....................................................................................... 4
Results 5
Incidence by Country. Norther Hemisphere 5
Time-course by country/region 7
Discussion 15
Strengths and shortcomings 15
Evidence for (at least) two different strains of SARS-CoV-2 15
Changes in infectivity and lethality between China and Europe 16
Predictions for COVID-19 in North America 17
A historical perspective 18
Implications for prevention 18
References 20
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -3-
Introduction
The first cases of a new coronavirus strain, termed SARS-CoV-2 (Severe Acute Respiratory Syn-
drome CoronaVirus) by the International Committee on Taxonomy of Viruses (ICTV) (Cascella 2020),
were reported on 31-12-2019 in Wuhan, the capital of the Hubei province of China.(Jernigan 2020) As
of 2020-03-28, 10:00 CET, 591,971 symptomatic cases and 27,090 deaths have been reported
from virtually every country in the northern hemisphere (see Section Data), The disease was
termed COVID-19 by the WHO on 2020-02-11, and categorized as a pandemic on 2020-03-12,
yet the details of the spread and their implications for prevention have not been discussed in
sufficient detail.
Between 02-14 and 03-16, the Dow Jones fell 31% from 29,440 to 20,188, raising fears for the
economy, in general, and retirement savings, in particular. Several administrations have imposed
severe restrictions aimed at containment of the virus. For instance,
On 03-08, the Italian government imposed a quarantine on 16 million people in the north of
Italy, which was followed up on 03-11 with a nationwide closure of all restaurants and bars
along with most stores.(WSJ, 2020-03-11)
on 03-11, the U.S. administration banned travel from 26 European countries (Austria, Belgium,
Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Italy,
Latvia, Liechtenstein, Lithuania, Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Slo-
vakia, Slovenia, Spain, Sweden, and Switzerland). https://www.cnn.com/world/live-news/coronavirus-outbreak-03-12-20-
intl-hnk/index.html.
On 03-15, New York’s Mayor de Blasio reversed his previous position that NY schools should
remain open to avoid health care workers from “staying home and watching their children”
and announced NY public schools to be closed, following many other school systems.
From 03-17, all New York, New Jersey, and Connecticut restaurants were had to close. From
03-22, hairdressers and barbers were also closed.
From 03-19 California was “shut down”.(Executive order N-33-20)
On 03-22, the lockdown in the Italian region of Lombardy was tightened to ban sports and
other physical activity, as well as the use of vending machines.
Also on 03-22, the National Guard was activated in New York, California, and Washington
State, five senators self-quarantined. NY governor Cuomo mandated that all nonessential
businesses close or work from home.
By the end on 03-20, the Dow Jones was down at 19,173 (35%) from 02-14. On 03-26, the U.S.
Senate approved a $2T stimulus package.
For most of the first three months of the epidemic, much of the response was driven by “fear,
stigma, or discrimination”(Ren 2020), including naming SARS-CoV-2 the “China virus”(Rogers 2020), de-
spite the fact that seasonal respiratory zoonotic pathogens typically originate in China, where life-
animal markets provide chances for animal viruses to transmit to humans.(Malik 2020)
After three months, enough data are available to discuss important epidemiological characteris-
tics of COVID-19 and the potential impact of interventions. In particular, we have now seen the
number of new cases (and deaths) to decline in China and South Korea and to at least stabilize
in some European countries. Changes in number of deaths follow the changes in number cases
(albeit at a lower level) by about two weeks. Hence, we can discuss both the infectiousness and
the lethality of the virus, two important characteristics to assess public health impact of the dis-
ease.
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -4-
One of the key findings published herein is evidence that at least two strains of SARS-CoV-2 with
different infectiousness and lethality have evolved, and by following the likely path for each of
these strains we can obtain novel insights into the nature of the epidemic and, thus, the effective-
ness of prevention strategies.
Materials and Methods
Data
All data were downloaded on 2020-03-27 from the European Centre for Disease Prevention and
Control (ECDC) Web site at https://www.ecdc.europa.eu/en/publications-data/download-todays-
data-geographic-distribution-covid-19-cases-worldwide, where data are collected daily between
6:00 and 10:00 CET. Updates were collected from the Johns Hopkins online tracker available at
https://systems.jhu.edu/research/public-health/ncov/.Population data were accessed from
https://www.worldometers.info/world-population/population-by-country/ on 2020-03-12. Data on
ages by country were accessed from https://data.worldbank.org/indicator/SP.POP.65UP.TO.ZS.
Methods and models
Statistical and Bioinformatics Methods
The data were processed with MS Excel. To avoid biases from inappropriate model assumptions
only basic descriptive statistics were employed. In some cases, data from only the day before or
after (or both) was averaged (up to a three day moving average) to reduce the effects of apparent
reporting artifacts (Darwin’s natura non facit saltum,(Berry 1985)) without creating undue biases. The
two “smoothers” applied were:
averaging x0 with a previous x1 (or, rarely, following x+1) data and
applying a moving average of (x1, x0, x+1)(2 x1 + x0, x1 + x0 + x+1, x0 +2 x+1)/3
No other changes were applied to the data.
Like China in mid-February, the German Robert-Koch-Institute (RKI) changed the reporting sys-
tem in mid-March, which resulted in a near 6-fold increase of the data reported on 03-20 over the
data reported on 03-19. Such changes in reporting systems add to the difficulties in interpreting
the data.
Epidemiological Models
If a disease causes immunity after an infectious period of a few days only, like respiratory dis-
eases, an epidemic extinguishes itself as the proportion of immune people increases. Under the
SIR model,(Kermack 1991) for a reproduction number(Dietz 1993) (secondary infections by direct contact
in a susceptible population) of R0=1.52.5 over 7 days (recovery rate: γ=1/7=.14), the noticeable
part of the epidemic lasts about 9045 days (R0/γ=β=.21.36) in a homogeneous population of
10M. The period is shorter for smaller more homogenous and longer for larger, more hetero-
genous populations. For a given infectious period 1/γ (here, e.g., 7 days. SARS and COVID-19
incubation period plus 2 days(Lauer 2020)), R0 also determines how long it will take for early cases to
become visible after a single import (15060 days), the peak prevalence of infections (522%),
and how many people will become immune (5590%). An arbitrary low rate of disease-related
death (0.00001*I/d) has been added to allow for comparison between models.
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -5-
Fig 1: SIR Model of SARS. Number of susceptible (blue), infectious (red), and resistant (green) people after a popu-
lation of 10,000,000 susceptible people is exposed to 20 subjects infected carrying a novel virus. Assumptions: R0 =
2.2, infectious period = 7 days,(available from https://app.box.com/s/pa446z1csxcvfksgi13oohjm3bjg86ql )
Results
Incidence by Country. Norther Hemisphere
Table 1 shows the raw daily incidence by population sizes for countries with epidemiological rel-
evance in the northern hemisphere. Countries within proximity are grouped by their peak inci-
dence (red background).
The Hubei province in China (with the capital Wuhan), South Korea, Iran, Italy (especially the
Lombardy region), and Spain have the highest peak incidence, followed by the countries neigh-
boring Italy.
It should be noted, however, that there is no uniform definition of “cases”. In some countries a
case needs to have symptoms, in other countries, it suffices to have antibodies (be immune).
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -6-
Table 1: Incidence by Country. Dates: Feb 13 (peak intensity in Mainland China, mostly Hubei and neighboring prov-
inces), Feb 19 to Mar 22. Countries with low population size (PSz) or low number of cases (Total) are hidden. Red
background indicates countries/dates with high incidence. Countries/regions are sorted by proximity among each other
and distance from Mainland China.
Country PSz [M] Tot al Total/ M 02-05
2-21
2-22
2-23
2-24
2-25
2-26
2-27
2-28
2-29
03-01 03- 02 03-03 03-04 03- 05 03-06 03- 07 03-08 03- 09 03-10 03- 11 03-12 03- 13 03-14 03- 15 03-16 03- 17 03-18 03-19 03- 20 03-21 03- 22 03-23 03- 24 03-25 03- 26 03-27 03- 28
CN(Hubei) 58.5 82,213 1405. 35 66.2 15 14 11 4978679. 81 3.5 2.17 2. 03 22.91 1.73 0. 79 0.77 0. 34 0.5 0.41 0. 38 0.32 0.38 0. 43 1.88 0.56 1. 28 1.69 1. 35 1.42 2.56 1. 69 1.69 2.07 1. 9 2.29
Vie tnam 97 169 1.74 0.01 000000000 0 0 0 0 0 00.01 0. 04 0.09 0. 01 0.04 0.04 0. 05 0.05 0.04 0. 04 0.04 0. 05 0.08 0.11 0. 04 0.11 0.19 0. 07 0.1 0.13 0. 11 0.11
Cambodia 17 102 6.00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.06 0000.06 0.06 0. 06 0.12 0. 29 0.71 0 0 0. 68 0.68 1.15 1. 15 0.12 0.2 0. 27 0.12 0. 24
South_Korea 51 9,478 185. 84 0.04 145335912 13 13.7 13.5 11. 8 10.1 9. 37 9.37 8.71 7. 18 5.64 3.66 3. 66 2.24 2. 16 2.1 1.49 1. 45 1.65 1.82 2. 56 2.52 2.49 1. 92 1.25 1. 49 1.96 2.04 2. 32 2.32
Singapore 6732 122.00 1 0010000100.67 0. 67 0.33 0. 33 0.33 1.28 1. 44 1.61 21.67 1 2 1.5 2. 17 2.33 22.83 3. 83 6.58 6.58 7. 06 6.11 5. 17 98. 17 2.56 9. 67 16.8
Malaysia 32 2, 161 67.53 0.06 000000000 00.06 0. 06 0.22 0. 44 0.52 0.52 0. 31 0.31 0.38 0. 44 0.45 0. 45 1.22 2.83 3. 71 4.58 3.75 3. 58 3.72 3. 85 4.47 5.08 5. 7 45. 34 6.69 4. 06
Japan 126 1,499 11. 90 0.04 0000000000.07 0.12 00. 11 0.39 0.33 0. 37 0.4 0.26 0. 21 0.43 0.4 0. 44 0.49 0. 34 0.27 0.08 0. 19 0.19 0.56 0. 46 0.36 0. 33 0.39 0.45 0. 6 0.76 1.07
Taiwan 24 267 11.13 0.04 0000000000.04 00. 04 0.04 00.08 0. 04 0000.13 00.04 0. 17 0.25 00.33 0. 42 0.86 0. 81 0.75 0.67 0. 83 10. 88 0.79 0. 71 0.63
Thailan d 70 1,136 16.23 0. 09 000000000 00.01 0 0 0.03 0. 03 0.02 0.02 0 0 0.13 0. 16 00. 17 0.23 0. 23 0.45 0.45 0. 25 0.25 2.58 2. 42 2.27 1. 52 1.54 1.57 1. 3 0
Indonesia 274 1,046 3. 82 0 000000000 00.01 00000000.02 0.02 0. 05 0.06 0.06 0. 09 0.09 0. 1 0.1 0.07 0. 17 0.27 0.37 0. 37 0.31 0.31 0. 38 0.44 0.5
Philippi nes 110 803 7.30 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.02 0.01 0. 04 0.21 00.15 0. 03 0.11 0.35 0. 35 0.21 0.21 0. 2 0.2 0.68 0. 68 0.25 0.52 0. 79 0.76 0. 76 0.76
India 1380 873 0.63 0 000000000 0 0 0 00. 02 00000.01 00. 02 00.01 0. 01 0.01 0.01 0. 01 0.01 0. 02 0.04 0.05 0. 06 0.06 0.05 0. 06 0.08 0. 09
Pakist an 221 1,197 5.42 0 0000000000.01 00000000. 02 0.02 00.02 0 0 0.04 0. 36 0.36 0.26 0. 26 0.56 0. 52 0.48 0.57 0. 52 0.47 0.47 0. 47 0
Afghani stan 39 91 2. 33 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.04 0.04 00.04 0. 04 0000.08 0. 15 0.13 0.03 0 0 0.05 0. 09 0.14 0.19 0. 45 0.45 0. 21 0.21
Sri_L anka 21 106 5.05 0000000000000000000000.05 0.05 0. 14 0.24 0.38 0. 48 0.41 0.38 0. 35 0.62 0. 57 0.43 0.48 0. 24 0.1 0.1 0
Iran 84 32,332 384.90 0 0000001122.44 4.58 6.23 8. 46 8.46 10.9 10. 9 10.8 10.8 8. 79 8.79 11. 4 12.8 15.3 15. 6 14.4 13. 2 14 14.2 13.2 12. 9 12.6 12. 2 16.8 21 26.3 28. 4 34.8
Iraq 40 458 11. 45 0 0000000000.15 0. 15 0.05 0. 13 0.13 0.18 00. 4 0.18 0 0 0. 23 0.1 0.28 0. 49 0.49 0. 38 0.38 0.29 0. 29 0.46 0.46 0. 59 0.85 1. 11 0.8 1.18 1. 57
Saudi_Arabia 35 1,104 31.54 0 000000000 00. 01 0.01 00.06 0. 06 0.03 0.03 0. 11 0.11 0. 14 0.71 0.49 0. 69 0.46 0.46 0. 43 0.36 11.64 1. 81 2.6 3.39 3. 66 3.66 3. 8 3.2 2.63
United_Arab_
10 405 40.50 0 000000010 0. 200.3 0. 3 0.1 0.1 0. 8 0.8 0.7 0. 7 1.5 0.55 0. 55 0 0 0.65 0. 65 11.4 1.8 0. 65 0.65 1.5 3.17 4.83 5. 67 5.23 4.8
Oman 5131 26.20 0 000000000 0 0 0.2 01.6 0. 2 0 0 0.2 0. 2 0 0 0. 2 0.2 00.4 0. 4 1111.3 1.3 0.6 2.2 3. 6 33.2 3.2
Kuwait 4225 56. 25 0 000112441 00. 25 1.25 1.25 0. 25 0.25 0.38 0. 38 0.75 0. 25 11. 17 2.58 4122. 75 2.17 2.08 23. 5 3.5 30.25 0. 5 1.75 2.83 3. 92
Qatar 3562 187.33 0 0000000000. 33 0.67 0. 83 0.83 00.5 0.5 0. 33 1 1 27. 8 27.1 26.4 19. 3 5.67 18.4 11. 7 4.89 3. 33 2.67 3.33 3. 67 4.33 4.33 4. 78 5.22 44.33
Bahrain 2466 233.00 0 0001111 503 1.5 1.5 1.5 11.5 0 2 3.5 87. 75 7.75 13 13 24 0.5 1.5 57978.33 9.67 14 10. 2 14.2 18.2 11. 8 11.8
Jordan 10 212 21.20 0 000000000 0 0 0.1 000000000000.7 0.7 1 1 1.05 1. 05 1.4 1.4 1. 5 1.3 1.5 2. 37 2.83 3.3
Israel 93, 035 337.22 0 000000000 00. 33 00.28 0. 28 0.22 0.22 0. 67 1.56 1.72 1. 72 1.33 1. 56 2.22 6.89 86. 93 6.78 6.63 15. 5 15.5 19 20. 9 41.2 54.2 43. 5 40.9 38.3
Lebanon 7391 55.86 0 0000000000. 14 0.86 0.43 0 0 0.43 0. 86 01. 43 1.29 02.86 0. 71 1.93 1. 93 0.86 1.5 1.5 1. 86 2.19 4. 62 7.05 2.57 2. 71 4.9 4.81 4. 71 3.29
Palestine 591 18. 20 0 000000000 0 0 0 0 0 1.4 1.8 0.6 00.2 1. 8 0.2 0.5 0. 5 0.6 00.2 0. 4 0.6 0. 6 0.5 0.5 1. 4 00.2 2.4 2. 4 1.4
Russia 146 1,036 7.10 0 000000000 0 00.01 0.01 0 0 0. 02 0.02 0 0 0. 05 0.05 0.06 0. 08 0.1 0.12 0. 12 0.17 0. 24 0.31 0.37 0. 55 0.72 00.39 1. 12 1.25 1.34
Azerbai jan 2165 82.50 0 000000001 0 1 0 00.75 0.75 1.5 0 0 0.5 0.5 1 0 1.5 1. 5 02. 25 2.25 4 4 2.25 2.25 5. 17 5.67 6.17 8. 75 8.75 21. 5
Georg ia 481 20.25 0 0000000000.25 00000.75 0. 75 0.75 0.25 0. 5 20. 25 0.25 1.25 0 0 0.75 0. 25 0.75 0.75 0. 75 1.5 1.25 1. 75 2.25 0. 75 1.5 0.5
Belaru s 994 10.44 0 000000000 0 00. 17 0.17 0. 22 00000.17 0.17 0. 33 0.5 0.5 0. 33 0.33 0.67 0. 7 0.74 1.22 1. 06 1.06 0.28 0. 28 0.28 0. 28 0.44 0.44
Italy 60 86, 498 1441.63 0 001122144 49.35 2. 43 11.1 9. 78 12.8 13 20. 8 24.9 30 23.7 33 42.3 42.5 52.7 52.7 62. 7 62.7 76. 3 86.2 96.1 104 93.9 84.1 87.3 92.3 97. 3 99.3
Switzerland 912,104 1344.89 0 0000000100.67 0.67 0.67 0. 78 2.22 3.33 13. 6 6.11 7.56 4. 67 12.9 16. 9 23.6 29.7 26. 4 46.7 46. 7 50 59.2 81.1 103 118 118 101 101 106 123 140
Austri a 97,697 855.22 0 0000000000.33 0.44 0.44 0. 67 0.56 1.33 3. 67 2.78 0.33 3. 22 5.67 7. 11 12.8 15.9 16. 8 21 25.1 29. 2 43.6 48.8 53. 9 50.3 68 85. 8 88.4 87.1 89. 4 91.7
Slovenia 2632 316.00 0 000000000 0 0 0 0 0. 5 2.5 1.5 1. 5 23. 75 3.75 13 19.5 22.5 20 19 17 11 11 11 16 16 15. 5 15.7 19 22.3 24.5 27.5
France 65 32, 964 507.14 0 0000000000. 66 0.46 0.74 0. 52 1.12 2. 12 2.92 3.16 4. 1 5.04 5.72 7. 65 9.15 12.1 12. 9 14.2 18. 6 18.5 22.4 26. 3 24.9 26.9 37. 1 47.4 40. 1 47.7 55.3 58. 6
Spain 47 64, 059 1362.96 0 0000000000. 68 0.36 0.66 0. 79 1.04 1.3 2. 4 1.19 8.23 8. 23 9.26 14. 5 14.5 26.1 32. 4 36.6 36.6 42. 3 54 73 75. 3 81 86.8 96.1 140 169 183 167
Portugal 10 4, 299 429.87 0 000000000 0 0.1 0. 1 0.2 0.1 0. 4 0.4 0. 8 0.9 0.9 1 1 1.9 3.4 5. 7 7.6 8.6 11. 7 17.7 19.1 23. 5 28 34.7 41. 3 41.2 49.5 57. 7 72.4
Germa ny 84 48,582 578.36 0 0000000000.64 0.21 0.33 0. 46 0.79 1.64 2. 9 1.99 1.08 2. 35 2.35 6.39 6. 39 8.25 8. 73 12.4 14 13. 2 32.2 51.3 48. 2 39 39.4 40. 4 40.4 59 68.8 74. 9
United_Kingd
68 14,543 213.87 0 0000000000.07 0.19 0. 06 0.16 0.5 0. 44 0.71 0. 63 0.99 0.71 0. 76 1.22 1.97 1. 72 6.37 3. 69 2.24 7.99 7. 99 9.95 9.95 12. 5 12.5 17. 6 17.6 21.4 31. 3 42.4
Poland 38 1, 389 36.55 0 000000000 0 0 00.03 00.05 0.05 0.03 0. 13 0.16 0.13 0. 24 0.47 0. 72 0.72 0.96 0. 96 1.45 1.45 1. 81 2.18 2. 56 2.58 3.03 3. 98 4.14 4.3 4. 42
Romania 19 1, 292 68.00 0 000000000 0 0 00.05 00.11 0. 05 0.32 0.11 0. 11 0.42 1. 05 11. 32 1.26 1. 37 2.16 2.12 2. 09 1.14 1.88 2. 61 4.82 6. 93 9.04 7.21 9. 3 11.4
Bulgaria 7293 41.86 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.29 0.29 00. 21 0.21 1. 71 1.71 1.43 1. 43 1.57 2.14 2. 14 2.29 3. 38 4.48 3.14 2. 43 2.71 33.14 4. 14
Netherlands 17 8,603 506. 06 0 0000000000.29 0. 35 0.29 0. 59 0.59 2.59 2. 98 3.59 4.2 3. 29 3.59 7.12 8. 08 8.94 9.8 10. 4 16.4 17. 2 20.4 24.1 33. 4 34.2 35 37. 3 43.3 49.3 59. 9 68.9
Belgium 12 7,284 607. 00 0 000000000 00. 08 0.5 0.42 0. 83 2.25 4.92 52. 58 3.25 2.33 3. 92 7.08 13.3 13. 6 13.6 14. 9 14.9 20.3 25. 8 38.5 47.3 41. 3 35.3 47. 8 69.2 90.7 87. 4
Czech_Repub
11 2,279 207.18 0 000000000 00. 27 0.18 00.27 0. 36 0.64 0. 64 0.55 0.73 2. 09 2.82 23.09 5. 82 6.48 6.67 6. 85 11.8 11. 8 13.7 13.7 11 11 17.5 25 32.6 19.7
Slovakia 2295 147.50 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.5 10.5 0. 5 11. 5 5.5 4. 5 78. 5 11.5 6.5 6. 5 6.5 11.5 10. 3 9.17 4.75 4. 75 619.8 19.8
Greece 10 966 96.60 0 000000000 0. 3 0000.3 2.2 1. 3 2.1 0. 7 1.1 0.6 0. 9 3.4 5.7 5. 97 5.4 4.83 3.37 3. 73 4.1 3.3 3. 3 9.4 7.1 6. 3 6.3 7. 1 7.4
Hungary 10 343 34.30 0 000000000 0 0 0 0.1 0.1 0.1 0. 1 0.3 0.1 0. 1 0.2 0.2 0. 6 0.6 0. 7 0.7 0.55 0.55 1. 75 1.75 2.3 2.3 3. 07 3.17 3.27 3.7 3. 7 4.3
Denmark 62, 046 341.00 0 0000000000. 17 0.17 0. 17 0.5 0.33 1. 67 0.5 1.33 1. 17 12.5 25.2 42 26.7 21.3 3.83 8.5 10. 9 13.4 15.2 11. 7 11.7 11. 8 11.3 14.7 18. 2 22.2 25.5 28. 2
Finland 61,025 170. 83 0 000000000 00.25 0.25 0.17 0. 42 0.42 1.17 0. 92 0.92 1.67 1. 58 1.58 8 8 9.17 6.61 6. 06 5.5 6.75 6. 75 8.33 11. 8 15.8 15.1 14. 3 14.7 13 11. 2
Sweden 10 3,046 304.60 0 000000011 0.1 0. 1 0.1 0.9 1. 1 2.6 5 5 4.2 4.5 7.8 13. 6 15.8 15. 5 14.9 10.8 8. 9 4.6 12.8 12. 8 16.2 16.2 14. 3 17.5 20.7 23. 8 29.6 24
Norway 53, 581 716.20 0 000000010 1.8 0. 8 1.2 1.6 4. 6 65.4 6.8 4. 4 4.6 25.5 28. 6 31.7 28.6 28. 6 28.8 26. 7 24.7 23.9 28. 9 33.9 38.3 41. 9 45.6 54. 5 54.5 66.5 66. 5
Icel and 2890 445.00 0 000000001 0 1 1.5 5 5 4.5 4515558810. 5 15.3 15.3 16. 5 21.8 27. 2 35.8 35.8 28. 8 28.8 37.3 37. 3 38.3 38. 3
Irel and 52,121 424.20 0 000000000 0. 2 0 0 0. 2 0.8 1.4 10. 2 0.4 1.4 1. 4 1.6 5. 4 4.2 7.6 810. 8 13.8 14.8 33. 9 27.9 22 24. 2 42.8 43.9 44. 9 55.7 55. 7
Croatia 4586 146. 50 0 000000001 0 0. 5 0.25 0. 25 00. 25 0.25 0.25 0 0 0.25 0. 75 2.25 1.5 1.5 2. 75 23. 25 35.75 12. 8 12.8 10.8 14. 7 18.6 12.4 17 21.6
Serbi a 9457 50.78 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.11 0 0 0.22 0. 22 1.06 1.06 1. 22 1.22 0. 61 0.61 1.67 2. 44 2.28 2.28 1. 56 4.33 3. 78 4688.11
No rth _Mace d
2219 109.50 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0.5 0.5 0110.5 0. 5 110365.5 7 7 11 11 9. 33 10.5 11. 7 10.5 10.5
Lithuania 3358 119.33 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.33 0.33 00. 5 0.5 1.33 1. 33 12.67 2.67 5 7 12 12.7 12 15.8 15.8 14 14
Latvia 2280 140. 00 0 000000000 00.25 0. 25 0000.25 0. 25 0.5 1.25 1. 25 2 2 1. 5 3 3 2. 5 8.75 8.75 10 10 7 7 16. 5 13.7 10.8 14. 8 14.8
Egypt 102 495 4.85 0 000000000 00.01 00000. 12 0.17 0.17 0. 06 0.04 0.1 0. 1 0.08 0.1 0. 11 0.16 0.39 0. 29 0.29 0. 29 0.22 0.23 0. 25 0.56 0.56 0. 26 0.26
Algeria 44 305 6.93 0 000000000 00.05 00.05 0.16 0.11 00. 03 0.03 0000. 09 0.05 0.25 0. 25 0.18 0.19 0. 2 0.23 0.16 0. 09 1.08 1. 08 0.89 0.88 0. 87 0
Morocco 37 345 9.32 0 000000000 00. 01 0.01 00.01 0. 01 00000.04 0.04 0. 03 0.03 0.3 0. 27 0.24 0.19 0. 27 0.43 0. 43 0.39 0.39 0. 51 0.97 1.44 1. 58 1.71
Tunisia 12 227 18. 92 0 000000000 00.04 0.04 000000.04 0.04 0. 25 0.17 0.5 0. 25 00.17 0.17 0. 33 0.42 0. 83 1111.17 2.08 3. 28 3.14 3
United_States
331 ###### 316.27 0 0000000000. 01 0.06 0. 04 0.07 0.1 0. 22 0.32 0.29 0. 37 0.6 0.82 0. 87 1.06 1.54 2. 35 2.49 2. 68 5.34 9.03 14. 6 16.2 21.5 25. 6 33.9 26. 6 42.2 50.7 56. 5
Mex ic o 129 717 5.56 0 0000000000. 02 0.01 000000. 01 0.01 00.02 0. 02 0.04 0.08 0. 12 0.09 0.22 0. 09 0.19 0. 36 0.3 0.37 0. 47 0.4 0.33 0. 54 0.85 1.02
Canada 38 4,689 123. 39 0.03 0000000000.11 0.11 0. 08 0.08 0.2 0. 2 0.16 0.16 0. 13 0.39 0. 42 0.48 0.73 0. 97 1.72 2.18 2. 63 3.61 3.7 3. 8 5.07 5.07 5. 24 5.68 8. 24 37.5 16.7 17. 7
Nigeria 206 24.21 0000000000000000000000000000.01 0. 02 0.01 0. 02 0.04 0.04 0. 04 0.03 0.05 0. 05 0.08
South_Africa 59 19.83 0 000000000 0 0 0 0 00.02 00.02 0. 02 0.07 00.1 0. 07 0.12 0.23 0. 23 0.19 0. 39 0.53 0.69 0. 7 0.71 1.11 1. 79 2.47 2.95 3. 46 3.98
Cameroon 27 3.26 0 000000000 0 0 0 0 00.02 0.02 0. 02 0.02 000000. 04 0.04 0.11 0. 11 0.07 0. 07 0.24 0.24 0. 54 0.54 0.59 0. 3 0.3 0
Senegal 17 7.00 0 000000000 00. 03 0.03 0.06 0. 06 0.06 0000000. 12 0.76 0.12 0. 29 0.06 0. 24 0.29 0.32 0. 32 0.53 0.65 0. 71 0.59 0. 59 0.59 0.59
Togo 83.13 0 000000000 0 0 0 0 00.06 0.06 0000000000000.5 0.5 0.38 0. 38 0.38 0. 31 0.31 0.13 0. 13
Australia 25 135.12 0.04 0000000000.04 0.12 0.16 0. 32 0.44 0.28 0. 16 0.44 0.24 0. 8 0.48 0.56 1. 2 1.79 1.89 23. 08 3.16 4.44 5. 76 6.6 16.7 16. 7 14.3 14. 3 14.9 14.9 8. 48
New_Ze aland
583.20 0 000000000 0 0 0.1 0.1 0.2 0. 2 0.2 0000000.2 0. 2 0.2 1.2 1. 2 1.6 2.2 2. 73 4.2 5.67 812.5 13. 1 13.6 15. 6
Brazil 212 16.12 0 000000000 0 0 0 0 00.02 0.02 00. 06 00.04 0.08 0. 12 0.1 0.2 0. 21 0.23 0.27 0. 65 0.91 1.24 1. 45 1.67 1. 63 1.46 1.49 1. 91 2.34
Colombia 51 10.57 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.02 00.02 0. 02 0.06 0.06 0. 07 0.07 0.35 0. 22 0.24 0. 44 0.44 0.54 0. 71 0.88 0.94 0. 94 1.54 1. 21 0.88 0.94
Argentina 45 15.33 0 000000000 0 00. 01 0.01 0.01 0. 01 0.13 0.02 0. 07 0.08 0. 08 0.13 0.13 0. 07 0.24 0.24 0. 2 0.31 0.4 0. 69 0.94 1.02 1. 1 1.16 1.75 2. 34 1.93 2. 24
Peru 33 19.24 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.03 0.09 0.09 0.06 0. 06 0.18 0. 15 0.48 0.15 0. 85 0.45 0.91 1. 49 2.08 1. 27 1.27 1.36 0. 97 1.86 1.87 1. 88 1.67
Chile 19 23184.74 0 000000000 0 0 00.05 0.11 0. 05 0.05 0.13 0. 13 0.16 0. 21 0.32 0.53 0. 53 0.84 0.84 3. 32 3.32 3. 71 3.71 5.13 5. 13 5 6 9. 26 11.6 12.3 12. 3
Ecuador 18 90.39 0 0000000000.06 0.28 0.06 0.08 0. 08 0.17 0 0 0. 06 0.06 0. 11 00. 17 0.17 0. 28 0.5 1.17 2. 94 3.17 5.35 6. 74 8.13 13.1 10. 2 7.3 8.33 10. 1 11.9
Dominican_Re
11 52. 82 0 000000000 00.09 00000.09 0 0 0.27 0000. 55 00000.45 0. 45 4.14 4.14 8. 18 3.91 6. 09 7.27 8.73 8. 45
Costa Ri ca 552.60 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.2 00. 8 0.8 0.8 1. 8 0.2 0. 6 0.9 0.9 1. 2 1.8 3.8 4. 13 3.2 2.27 3. 4 4.47 4.47 4.47 66.4
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -7-
Time-course by country/region
Among the Hubei population of 58.5M, the incidence rose from the first case reported in late 2019
to about 60 new cases per million people per day by 02-05 and then steadily declined (Fig 1) from
~4000 on 02-05 to below 50 cases per day since 03-08.
Fig 2: COVID-19 cases in Mainland China. Blue: cases/M/d, red: deaths/M/d. Around Feb 13, the case definition was
expanded, resulting in additional cases from previous days being added. Hence, the 02-13 cases have been truncated.
Most cases were seen in the Hubei province of 58.5M people (see Table 1 for population sizes).
By Mid-January 2020, the first cases of COVID-19 were seen in other Asian countries, but inci-
dence remained below about 1/M/d outside of continental China, except for an increase to about
2.5/M/d in Malaysia/Brunei and 10/M/d in Singapore, but including Japan with the largest propor-
tion of people 65 years of age and older in the world (28%).
Fig 3: COVID-19 cases in Maritime Asia. See Fig 1 for legend.
In continental South Korea (population 51M, cumulative incidence 185/M), however, the incidence
soon rose to a peak of about 14/M/d between 02-29 and 03-02, before declining to less than 150
cases per day (3/M/d) since 03-12 (Fig 2a).
In Iran (cumulative incidence 385/M), incidence rose about a week after South Korea. The top
incidence before 03-23 (~15.5 cases/M/d) was about the same (the recent increase on 03-24..28
may indicate a “rebound” into a population not immunized by the previous wave(s)). Lethality in
Iran was notably higher and followed the increase in cases with a delay of several days (Fig 2b)
as also seen in South Korea (Fig 2a, 03-06 for deaths vs 02-28..03-01 for cases).
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
01-15
01-16
01-17
01-18
01-19
01-20
01-21
01-22
01-23
01-24
01-25
01-26
01-27
01-28
01-29
01-30
01-31
02-01
02-02
02-03
02-04
02-05
02-06
02-07
02-08
02-09
02-10
02-11
02-12
02-13
02-14
02-15
02-16
02-17
02-18
02-19
02-20
02-21
02-22
02-23
02-24
02-25
02-26
02-27
02-28
02-29
China
0
0.2
0.4
0.6
0.8
1
1.2
02-21
02-23
02-25
02-27
02-29
03-02
03-04
03-06
03-08
03-10
03-12
03-14
03-16
03-18
03-20
03-22
03-24
03-26
03-28
Maritime Asia
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -8-
Fig 4: COVID-19 cases in South Korea and Iran. See Fig 1 for legend.
From 03-19 to 03-20, several European countries have seen a more than two-fold increase in the
number of cases reported (Germany:570%, San Marino:340%, Ireland:260%, Switzerland:
240%, Austria: 202%). As natura non facit saltum (Darwin: nature doesn’t jump),(Berry 1985) such
abrupt increases must be, at least in part, the result of reporting or other artifacts. In Germany, for
instance, the reporting system was changed between 03-16 and 03-19, so that the number is
likely includes cases previously reported only through a parallel system. France, Italy, and Spain
also reported an unusual increase by 2735 percent. All these countries reported lower in the
following days. As more data are reported, some averaging has been applied.
0
2
4
6
8
10
12
14
16
02-21
02-23
02-25
02-27
02-29
03-02
03-04
03-06
03-08
03-10
03-12
03-14
03-16
03-18
03-20
03-22
03-24
03-26
03-28
South_Korea
0
5
10
15
20
25
30
35
40
02-21
02-23
02-25
02-27
02-29
03-02
03-04
03-06
03-08
03-10
03-12
03-14
03-16
03-18
03-20
03-22
03-24
03-26
03-28
Iran
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -9-
Fig 5: COVID-19 cases in Italy and it’s neighboring countries. Italy (top), European countries neighboring Italy (IT+,
middle. Spain also shown separately. See Fig 1 for legend.
Among European countries with a population of more than 2M, Italy has the highest cumulative
incidence per capita (Table 1, Fig 3a), followed by its neighbors Spain and Switzerland (included
in Fig 3b). In all three European countries, the cumulative incidence is now similar to that in the
Hubei province (their population of 5060 M is also similar).
0
20
40
60
80
100
120
02-21
02-23
02-25
02-27
02-29
03-02
03-04
03-06
03-08
03-10
03-12
03-14
03-16
03-18
03-20
03-22
03-24
03-26
03-28
IT Neighbors (CH/F/ES/AT/SI )
0
20
40
60
80
100
120
140
160
180
200
02-21
02-23
02-25
02-27
02-29
03-02
03-04
03-06
03-08
03-10
03-12
03-14
03-16
03-18
03-20
03-22
03-24
03-26
03-28
Spain
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -10-
The overall epidemic in Europe (Fig 4a) is a population weighted average of the high incidence
regions (Fig 4a, weighted average of Fig 3) and the remaining low incidence countries (Fig 4c).
Fig 6: COVID-19 cases in Europe. Early onset/high lethality (IT and neighbors, top), total (middle), and late onset/low
letality (other European countries, bottom). See Fig 1 for legend.
The incidence in the countries with early onset and high lethality (Italy and its neighbors, including
Spain and France) now seems to be leveling off, after about 4 weeks from 1/M (Italy: 02-26..~03-
22, neighbors: 03-01..~04:01), compared to Hubei’s and South Korea’s 2 weeks (01-19..02-05
0
20
40
60
80
100
120
02-21
02-23
02-25
02-27
02-29
03-02
03-04
03-06
03-08
03-10
03-12
03-14
03-16
03-18
03-20
03-22
03-24
03-26
03-28
EU/high
0
10
20
30
40
50
60
70
02-21
02-23
02-25
02-27
02-29
03-02
03-04
03-06
03-08
03-10
03-12
03-14
03-16
03-18
03-20
03-22
03-24
03-26
03-28
Europe
0
5
10
15
20
25
30
35
40
45
02-21
02-23
02-25
02-27
02-29
03-02
03-04
03-06
03-08
03-10
03-12
03-14
03-16
03-18
03-20
03-22
03-24
03-26
03-28
EU/low
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -11-
and 02-21..03-06). Other European countries, where the epidemic started more slowly, may see
the peak in early April. Germany is notable for reporting 48,582 cases, but only 325 deaths (03-
28).
Fig 7: COVID-19 cases in selected European countries. See Fig 1 for legend. Data in Germany is based on cases
reported electronically to the Robert-Koch-Institute (RKI) and transmitted to the ECDC, but the RKI also provides two
sources of data on its Web site that are difficult to reconcile with these data.
0
10
20
30
40
50
60
70
80
02-21
02-23
02-25
02-27
02-29
03-02
03-04
03-06
03-08
03-10
03-12
03-14
03-16
03-18
03-20
03-22
03-24
03-26
03-28
Germany
0
10
20
30
40
50
60
70
02-21
02-23
02-25
02-27
02-29
03-02
03-04
03-06
03-08
03-10
03-12
03-14
03-16
03-18
03-20
03-22
03-24
03-26
03-28
France
0
10
20
30
40
50
60
70
80
90
100
02-21
02-23
02-25
02-27
02-29
03-02
03-04
03-06
03-08
03-10
03-12
03-14
03-16
03-18
03-20
03-22
03-24
03-26
03-28
Austria
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -12-
The epidemic in North America started later, especially in the US (except for a few isolated cases
likely imported directly from Asia). The incidence is still lower than in the older European popula-
tion, but keeps rising, as “more and more states are reporting” (https://www.cdc.gov/coronavirus/2019-ncov/cases-
updates/summary.html, accessed on 03-18). The increase seen is consistent with the dynamics of an emerging
epidemic. Cumulative incidence in the US (260/M) is still about half of that in Europe (507/M),
which reflects, at least in part, the later onset of the epidemic in the US.
Fig 8: COVID-19 cases in the North America. See Fig 1 for legend. The high peak in the Canadian data on 03-26
cannot be reconciled with the dynamics of a respiratory disease spreading.
A Widow of Opportunity for Containment (Social Distancing)
The effect of reducing the reproduction number by reducing the number of contacts (“contain-
ment”, “Social Distancing”) depends on when it starts in the course of the epidemic. Fig 10 shows
the effect of a one- month intervention cutting R0 in half starting at the point of the peak prevalence
of infectious subjects. Compared to Fig 1, the duration of the epidemic is shortened, albeit at the
price of reducing the R/S ratio, so that a subsequent epidemic with the same are similar virus
(cross-immunity) could start earlier.
0
5
10
15
20
25
30
35
40
02-21
02-23
02-25
02-27
02-29
03-02
03-04
03-06
03-08
03-10
03-12
03-14
03-16
03-18
03-20
03-22
03-24
03-26
03-28
Canada
0
10
20
30
40
50
60
02-21
02-23
02-25
02-27
02-29
03-02
03-04
03-06
03-08
03-10
03-12
03-14
03-16
03-18
03-20
03-22
03-24
03-26
03-28
United_States_of_America
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -13-
Fig 9: SIR Model of SARS, Window of Opportunity for Fast Eradication of the Epidemic. (see Fig 1 for legend).
The gray area indicates the period where containment can give a “coup de grace” to a respiratory disease epidemic.
The more narrow bell curve with a post-peak intervention indicates the reduction in number of infections and, thus,
deaths.(spreadsheet for model calculations available from https://app.box.com/s/pa446z1csxcvfksgi13oohjm3bjg86ql )
Fig 11 shows a one-month intervention starting about two weeks earlier, at the turning point where
the curve of the new cases changes from increasing faster to increasing more slowly. This inter-
vention reduces the number of deaths, but the epidemic is extinguished two months later and the
R/S ratio (“herd immunity”) is further decreased.
Fig 10: SIR Model of SARS, Window of Opportunity for Maximal Reduction of Total Deaths. (see Fig 1 for legend).
The gray area indicates the period where containment can have the most impact on total number of deaths. However,
the epidemic is not eradicated.(spreadsheet for model calculations available from https://app.box.com/s/pa446z1csxcvfksgi13oohjm3bjg86ql )
Fig 12 shows the effect of an intervention that starts even earlier, about two weeks before the
intervention in Fig 11. Even if the intervention is extended from one to four months no herd im-
munity is created and, thus, the epidemic rebounds.
`
``
`
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -14-
Fig 11 : SIR Model of SARS, Effect of Early Social Distancing / Lockdown. (see Fig 1 for legend). It is assumed
that a highly effective intervention reduces R0 by 50% for 4 months, beginning after the appearance of a novel type of
cases is noticed. The death rate of 5K/10M is arbitrary.(spreadsheet for model calculations available from
https://app.box.com/s/pa446z1csxcvfksgi13oohjm3bjg86ql )
In summary, there is a narrow window-of-opportunity for interventions (“flattening the curve”) aim-
ing to improve public health by reducing R0:
Starting after the peak prevalence (of infections) has little effect (not shown). The curve
goes down, but is not “flattened”.
Starting at the peak prevalence gives the epidemic a “coup de grace”, shortening its du-
ration, albeit at the price of reducing the R/S ratio. The curve is narrower, but not “flat-
tened”.
Starting at the peak incidence “flattens” the curve without broadening it and maximizes
the number of deaths prevented during the current epidemic, but reduces herd immunity
and, thus, the chance of another epidemic coming sooner.
Starting before the peak incidence “flattens the curve”, but also broadens it and causes
a rebound, unless the intervention is continued for many more months.
It is herd immunity that stops the spread of an infectious disease, so in general, one would want
to let the epidemic initially run its natural course (or even accelerate it, as people have traditionally
done with “measles parties”) to build immunity as fast as possible.
To reduce the duration of the epidemic and its impact on the economy (and also increase the time
until the next epidemic can spread), one would wait until the prevalence of infectious people (I)
reaches its peak (in the above model: day 83, red).
Without repeated broad testing, however, this date cannot be directly observed, but it is known
that peak prevalence of infected people is followed about a week by peak number of new cases.
This is too late to make a decision, but the SIR model shows that this peak preceded by two
weeks by the turning pointin cases where the curve of the new cases changes from increasing
faster to increasing more slowly (day 76), which can be estimated from the observed cases in
time to making a decision. (It is also about 50% of the peak number of new cases, which one
might be able to predict.) Hence, peak prevalence (of infections) follows the turning point/half
peak (in number of cases) by about a week. The window of opportunity for starting an inter-
vention is the week following the turning point in number of cases per day.
`
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -15-
Discussion
Strengths and shortcomings
A major strength of this analysis of the epidemiological data is that it does not rely on epidemio-
logical models with questionable assumptions. Instead, the results reflect raw incidences over
time as reported by the ECDC, depicted by country or region of neighboring countries.
A shortcoming of such an entirely data-based approach is that it lacks the sophistication and
potential additional insights that could come from fitting, e.g., differential equation models. The
only modeling assumption made is that curves should be “smooth” (except when reporting arti-
facts are suspected), but even then, data were redistributed only to the directly neighboring day.
Still, the evidence is strong enough to draw qualitative conclusions about possible scenarios for
the spread of SARS-CoV-2 in the near future. Also, the results suggest strategies to explore the
variability of the SARS-CoV-2 virus strains and to select prevention strategies.
Evidence for (at least) two different strains of SARS-CoV-2
During the 2003 SARS epidemic the number of new cases peaked about three weeks after the
initial increase of cases was noticed and then declined by 90% within a month. Table 1 shows the
relevant timepoints for the 2020 COVID-19 epidemic. The SARS-C0V-2 data also suggest that it
takes at least a month from the first case entering the country (typically followed by others) for the
epidemic to be detected, about three weeks for the number of cases to peak and a month for the
epidemic to “resolve”. This data is consistent with the results from the SIR model (see Epidemio-
logical Models).
Table 2: Epidemiological Timepoints by Country Top : 2003 SARS, Bottom: 2020 SARS-CoV-2
First
cases
Gap
[d]
Begin
(>.1/1M)
Gap
[d]
Peak
Gap
[d]
End
(<.1/1M)
Begin to
End [d]
CA/SGH
02-23
03-02
18
03-20
28
04-18
28
(Svoboda 2004)
CA/NYG
04-20
04-27
30
05-27
6
06-03
36
(Svoboda 2004)
Guangdong
2002 (10)
>50
01-19
22
02-11
81
05-02
103
(Cao 2019)
Shanxi
03-15
34
04-19
30
05-19
64
(Cao 2019)
Beijing
03-05
49
04-24
30
05-24
69
(Zhou 2003; Cao 2019)
Mongolia
03-22
22
04-14
40
05-24
62
(Cao 2019)
Hebei
04-04
20
04-24
24
05-19
44
(Cao 2019)
Tianjin
04-14
8
04-22
12
05-04
20
(Cao 2019)
HK/PWH
02-21
03-11
6
03-17
HK/AG
03-20
4
03-24
29
04-13
(Zhou 2003; Leung 2004)
HK/
04-12
54
06-06
Taiw a n
02-24
23
03-17
38
04-25
50
06-14
88
(Small 2003; Yeh 2004)
Singapore
03-15
20
04-05
24
04-29
44
(Small 2003; Goh 2006)
Vietnam
02-23
11
03-04
28
04-06
39
(Shi 2003)
Median 2003
03-15
20
04-13
29
05-19
44
China (Hubei)
2019 (27)
>50
01-17
18
02-05
27
03-18
60
0 in Hubei
S- Korea
01-20
31
02-19
12
03-01
>25
>03-25
>37
Ongoing low level
Iran
02-20
2
02-22
14
03-07
Several waves?
Italy
01-31
22
02-22
30
03-22
2nd wave?
Germany
01-28
30
02-27
France
01-25
31
02-27
>30
?04-01
~04-15?
US
01-25
40
03-05
>30
?04-o7
~05-01?
Median 2020
31
02-22
25
03-25
~04-15?
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -16-
The 2003 SARS and the 2020 SARS-CoV-2 are not only similar with respect to genetics (79%
homology),(Lu 2020) immunology,(Ahmed 2020) involvement of endocytosis (also with influenza and syn-
cytial viruses),(Behzadi 2019) seasonal variation (same season in the northern hemisphere also with
influenza, syncytial, and metapneumo viruses)(Olofsson 2011), evolution (origin in bats, 88% homol-
ogy),(Benvenuto 2020; Malik 2020) but also with respect to the duration between emergence and peak of
cases as well as between this peak and resolution of the epidemic (Table 1). Based on these
similarities, one could predict the COVID-19 epidemic to end before 04-15 in Europe and about
two weeks later in the U.S.
The time and height of the peak incidence if cases in the different countries are consistent with
the hypothesis that SARS-CoV-2 moved step-by-step westward from China, via other Asian coun-
tries to Middle East (Iran. Qatar, and Bahrain), Southern Europe (Italy, followed by its neighbors
CH/F/ES/AT/SI), central and northern Europe, and, finally, the US.
Viruses improve their “survival” if they develop strategies to coexist with the (human) host.(Woolhouse
2007) Multiple coronaviruses have been found to coexist in bat populations.(Ge 2016) The emerging
COVID-19 data is consistent with the hypothesis that (at least) two SARS-CoV-2 strains have
developed. One strain, which traveled through South Korea, remained more infectious, while the
other strain, which traveled through other Asian countries lost more of its infectiousness. The
strain that passing through South Korea and then Iran and Italy (SKII strain) showed high lethality
in Iran and Italy, but less lethality when it traveled to Italy’s neighbors, either because of differ-
ences in health systems, because the strain mutated back, or because a strain arriving directly
from Asia had the advantage of spreading first. Only sequencing samples from these countries
can help to answer these questions.
Fig 12: Hypothetical virus transmission pathways. Connection width: number of contacts, box colors: infectious-
ness, box borders: lethality, dotted connections/borders: unknown. The end date of a box indicates the date of peak
incidence, if known (bold date).
Changes in infectivity and lethality between China and Europe
Mainland China is not reporting relevant numbers of novel cases anymore and Hubei reports no
new cases since 03-19. The number of new cases in South Korea also has declined to low levels
since its peak around 02-30. Maritime Southeast Asia continues to show low levels of new cases
only (<2.5/M/d), with the possible exception of Singapore and Malaysia/Brunei.
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -17-
The data are consistent with the same “SKII” strain traveling from China via South Korea and Iran
to Italy. Iran was hit about a week after South Korea (around 03-07), with a similar peak incidence,
but higher lethality (red bars in Fig 2). The data also suggests a second wave of infections in Iran,
which may have peaked on 03-15. Italy was hit a week after the first wave in Iran (which peaked
around 03-07/08, Fig 4a). Incidence in Italy reached a substantially higher incidence than reported
in Iran. The peak incidence in Italy may have been reached on 03-22 (at about 100/M/d).
Without sufficiently detailed genetic data, it is not clear whether the high lethality in Italy is due to
genetic variations in the virus or to Italy having the second oldest populations in the world (after
Japan). A 03-20 report by the Istituto Superiore di Sanita,(COVID-19 Surveillance Group 2020) however, impli-
cates that age and comorbidities played a role among 3200 deaths, mostly in Lombardy and
Emilia-Romana, median age was 80 years (IQR 73-85, only were 36 below the age of 50), 98.8%
had at least one comorbidity (hypertension: 74%, diabetes: 34%, ischemic heart disease: 30%,
atrial fibrillation, 22%, chronic renal failure: 20%, …).
The epidemiological data does not support the hypothesis that SARS-CoV-2 spread from Munich
in Germany to Italy.(Kupferschmidt 2020) Instead, the virus may have spread from Italy to its neighboring
countries, Switzerland, France, Spain, Austria, and Slovenia, within just a few days of arriving
from Iran. The top incidence seems to be less than half of that in Italy and the lethality is lower,
too. While Italy has many people 65 years and older (23%, second only to Japan data.worldbank.org/indi-
cator/SP.POP.65UP.TO.ZS), the relatively small differences or age distribution within Europe (e.g., Ger-
many: 21%) are unlikely to account for much of this difference. A possible explanation (indicated
in Fig 6) is that the less virulent strain(s) arriving from other parts of Asia may have had a head
start in those countries, so that imported infections from Italy met subjects who had already de-
veloped (cross) resistance against both strains.
Infections in Scandinavia arrived yet another half week later, but peaked around 03-12. The recent
data may indicate a second wave more similar to other European countries. The parts of Europe
not hit by the SKII strain may remain at much lower levels (below 30/M/d, except for effects of the
recent changes in the reporting systems). Overall, incidence in Europe seems to be leveling off.
After the effects of the changes I the reporting system have ceased, incidence in Europe may
peak soon at about 40/M/d.
Predictions for COVID-19 in North America
From Table 1, SARS-CoV-2 has arrived in the U.S. almost a week after it arrived in Europe. The
incidence is still low (currently at about 55/M/d) and is likely to continue to increase until early
April. If incidence in the U.S. were to peak at about 75/M/d, as in Europe as a whole (Fig 4b), one
would expect new cases to peak at up to 25,000 per day and the cumulative incidence could
reach 600/M (3 times the number of cases per million people in South Korea to account for a
longer course because of the size of the countries) or a total of 200,000 cases, and, at 2% lethality,
about 4,000 deaths, about four times the currently reported cumulative number of 1,079. These
are conservative estimates, because only 1% of cases died in South Korea over the course of the
epidemic and both countries have a similar proportion of people older than 65 years (14% vs
16%) On the other hand, the numbers could double if the SKII strain should have hit the US
earlier. A number of 4000–8000 U.S. death over the course of the epidemic compares to an ex-
pected number of 16,00078,000 influenza deaths per season from pneumonia and respira-
tory/circulatory complications alone, which also occur predominantly among people at 65 years
of age and older.(Rolfes 2018).
The precise number of people dying depends on (a) which virus strain got to the US first and (b)
how early people are being treated against severe complications (e.g., pneumonia).
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -18-
A historical perspective
This is not the first, and likely not the last time, that well-intentioned public health policies are
inconsistent with our understanding of how epidemics spread. For instance, during much of the
HIV epidemic, there was widespread fear that HIV could establish itself in the population as a
whole, even though the data (including data showing absence of transmission to the wives of
hemophiliacs)(Wittkowski 1995a) and models(Wittkowski 1992; Seydel 1994) contradicted this fear.(Wittkowski 1995b; 1996)
These results have been repeatedly confirmed.(Centers for Disease Control and Prevention 2019; H addad 2019) In the
case of heterosexual transmission of HIV one could argue that there was little risk associated with
a the public health policy promoting condom use, but in the case of COVID-19 prevention, ignoring
models and data may carry substantial risk.
During the AIDS epidemic, epidemiologists had the advantage that, in addition to the date of re-
port, the date of diagnosis was available for analysis so that variations in reporting delays, such
as mid-February in China, 03-20 in Germany, and 03-26 in Canada, could be accounted for. Un-
fortunately, the public COVID-19 data lacks that information.
Implications for prevention
A major problem with respiratory diseases is that one cannot stop all chains of infections within
families, friends, neighbors, … . Even after a couple of weeks of “lockdown” there will be a few
infectuous persons, and as long as there are enough susceptible people in the society, this is
enough to re-start the epidemic until there are enough immune people in the society to create
“herd immunity”. Hence, one would expect the cases to appear in waves (Fig 12, the period of
the “lockdown” corresponds to March to May, 2020 in the U.S.). Such waves of cases have been
seen in different countries and the longer than expected duration of the epidemic supports the
hypothesis that the social distancing / lockdown interventions had some effect, albeit at a high
cost for approx. 10% of deaths saved.
This analysis of the publicly available data suggests that at the time Italy imposed quarantine on
the Lombardy and adjacent regions on 03-08, the SKII virus strain had already reached the adja-
cent countries (Switzerland, France, Spain, Austria, Slovenia). Even though the lockdown started
early (03-08), which may have caused a rebound consistent with a decline in compliance.
In the US, the growth in reported cases per day slowed down after 03-20, yet there is no sign of
a turning point, yet. Still, New York, New Jersey, and Connecticut restaurants were ordered to
closed from 03-20; the shutdown of California was ordered on 03-19 (Executive order N-33-20). As social
distancing was ordered before the epidemic reached its turning point, a “flattened curve” is to be
expected, but the curve will also be broader.
Some containment strategies could even be counterproductive in other ways. For instance, the
simple model used in Fig 12 does not account for age-stratification. In diseases such as COVID-
19, where children develop mostly mild forms, while elderly people have a high risk of dying.(Zim-
mermann Curtis 2020) Hence, containment of high-risk groups, like elderly people in nursing homes (see
the Washington State example) is highly effective in protecting them from becoming infected and
reducing the pool that would have to reach herd immunity. A substantial increase in the duration
of the epidemic, however, might make effective containment of the elderly more difficult and, thus,
increase the number of deaths among the elderly.
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -19-
In the U.S. as a whole, the “turning point” for cases cannot be earlier than 03-25, but the New
York Times reports that New York and Detroit reached the turning point on 03-19. Hence, the
optimal time point for starting a New York public health intervention to reduce duration and impact
of the COVID-19 epidemic was around 03-27. Social segregation against COVID-19 in NY, as
one of the epicenters of the U.S. epidemic, started about 03-17 (day 73) with restaurants being
closed, and intensified on 03-22 (day 78) with all non-essential businesses being closed. Because
of the Easter holiday, restrictions are discussed to be lifted on 04-12 (day 100). The model predicts
that such an intervention would reduce the number of deaths in New York and Detroit (and pos-
sibly some other parts of the U.S., but the virus would linger on for another two months, so that it
would not be safe for (elderly) high-risk people to participate in this year’s Easter activities.
Fig 13: SIR Model of SARS, Phased in Restrictions. (see Fig 1 for legend). The gray areas indicates the periods of
low and high intensity restrictions).(spreadsheet for model calculations available from https://app.box.com/s/pa446z1csxcvfksgi13oohjm3bjg86ql )
Conclusions
Until a vaccine will become available, the only pharmacological strategy to reduce the number of
deaths is to reduce the damage the infection (and immune system) does, e.g., by reducing the
initial viral load,(Chu 2004) and making sure that people get treated at the earliest signs of pneumonia.
Aside from separating susceptible populations (elderly and high-risk subjects, e.g., in nursing
homes) from the epidemic, which is effective as long as virus is circulating, public health interven-
tion aiming to contain a respiratory disease need to start within a narrow window of opportunity
starting at or a week after the curve of the new cases changes from increasing faster to increasing
more slowly. Unless the containment efforts started earlier and prevented the epidemic from gen-
erating a sufficient number of immune people, the containment efforts can cease after about a
month or two (depending on late or early start, respectively), when the ratio of infectious vs im-
mune people is too low to for the disease to rebound. When the window of opportunity has been
missed, containment has only limited impact on the course of the epidemic.
To determine that time point, case data collected and reported needs to contain not only the date
of report, but also the date of “diagnosis” and whether the patient had clinical symptoms or was
merely tested positive and whether the patient was positive for circulating virus RNA/DNA (cur-
rently infectious) or antibodies (already immune).
``
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -20-
References
Ahmed SF, Quadeer AA, et al. (2020). Preliminary Identification of Potential Vaccine Targets for the
COVID-19 Coronavirus (SARS-CoV-2) Based on SARS-CoV Immunological Studies. Viruses 12(3).
Behzadi MA, Leyva-Grado VH (2019). Overview of Current Therapeutics and Novel Candidates Against
Influenza, Respiratory Syncytial Virus, and Middle East Respiratory Syndrome Coronavirus Infections.
Frontiers in microbiology 10: 1327.
Benvenuto D, Giovanetti M, et al. (2020). The global spread of 2019-nCoV: a molecular evolutionary
analysis. Pathog Glob Health: 1-4.
Berry RJ (1985). Natura non facit saltum. Biological Journal of the Linnean Society 26(4): 301-5.
Cao W, Fang L, et al. (2019). What we have learnt from the SARS epdemics in mainland China? Global
Health Journal 3(3): 55-9.
Cascella M, Rajnik M, et al. (2020). Features, Evaluation and Treatment Coronavirus (COVID-19).
StatPearls. Treasure Island (FL), StatPearls Publishing.
Centers for Disease Control and Prevention (2019). Estimated HIV incidence and prevalence in the
United States, 2010-2016. HIV Surveillance Supplemental Report 24(1).
Chu CM, Poon LL, et al. (2004). Initial viral load and the outcomes of SARS. Cmaj 171(11): 1349-52.
COVID-19 Surveillance Group (2020). Characteristics of COVID-19 patients dying in Italy Report based
on available data on March 20th, 2020. Rome, Italy, Istituto Superiore di Sanita.
Dietz K (1993). The estimation of the basic reproduction number for infectious diseases. Stat Methods
Med Res 2(1): 23-41.
Ge XY, Wang N, et al. (2016). Coexistence of multiple coronaviruses in several bat colonies in an
abandoned mineshaft. Virol Sin 31(1): 31-40.
Goh KT, Cutter J, et al. (2006). Epidemiology and control of SARS in Singapore. Ann Acad Med
Singapore 35(5): 301-16.
Haddad N, Robert A, et al. (2019). HIV in Canada-Surveillance Report, 2018. Can Commun Dis Rep
45(12): 304-12.
Jernigan DB (2020). Update: Public Health Response to the Coronavirus Disease 2019 Outbreak - United
States, February 24, 2020. MMWR Morb Mortal Wkly Rep 69(8): 216-9.
Kermack WO, McKendrick AG (1991). Contributions to the mathematical theory of epidemics--I. 1927.
Bull Math Biol 53(1-2): 33-55.
Kupferschmidt K (2020). Genome analyses help track coronavirus' moves. Science 367(6483): 1176-7.
Lauer SA, Grantz KH, et al. (2020). The Incubation Period of Coronavirus Disease 2019 (COVID-19)
From Publicly Reported Confirmed Cases: Estimation and Application. Ann Intern Med.
Leung GM, Hedley AJ, et al. (2004). The epidemiology of severe acute respiratory syndrome in the 2003
Hong Kong epidemic: an analysis of all 1755 patients. Ann Intern Med 141(9): 662-73.
Lu R, Zhao X, et al. (2020). Genomic characterisation and epidemiology of 2019 novel coronavirus:
implications for virus origins and receptor binding. Lancet 395(10224): 565-74.
Malik YS, Sircar S, et al. (2020). Emerging novel coronavirus (2019-nCoV)-current scenario, evolutionary
perspective based on genome analysis and recent developments. Vet Q 40(1): 68-76.
Olofsson S, Brittain-Long R, et al. (2011). PCR for detection of respiratory viruses: seasonal variations of
virus infections. Expert Rev Anti Infect Ther 9(8): 615-26.
Ren SY, Gao RD, et al. (2020). Fear can be more harmful than the severe acute respiratory syndrome
coronavirus 2 in controlling the corona virus disease 2019 epidemic. World J Clin Cases 8(4): 652-7.
Rogers K, Jakes L, et al. (2020). Trump Calls It the ‘Chinese Virus.’ Critics Say That’s Racist and
Provocative. New York Times. New York, A.G. Sulzberger: 11.
Rolfes MA, Foppa IM, et al. (2018). Annual estimates of the burden of seasonal influenza in the United
States: A tool for strengthening influenza surveillance and preparedness. Influenza Other Respir
Viruses 12(1): 132-7.
Seydel J, Kramer A, et al. (1994). Backcalculation of the number infected with human immunodeficiency
virus in Germany. J Acquir Immune Defic Syndr 7(1): 74-8.
Shi Y (2003). Stochastic dynamic model of SARS spreading. Chinese Science Bulletin 48: 1287-92.
Small M, Shi P, et al. (2003). Plausible models for propagation of the SARS virus. Trans Fundam
Electron, Commun Comput Sci E87-A.
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
Knut M. Wittkowski: Two epidemics of COVID-19 (2020-03-28) -21-
Svoboda T, Henry B, et al. (2004). Public health measures to control the spread of the severe acute
respiratory syndrome during the outbreak in Toronto. N Engl J Med 350(23): 2352-61.
Wittkowski KM (1992). Estimates of HIV prevalence in Germany and the U.S. Sensitivity to assumptions
concerning reporting delays and incubation time. VIII International Conference on AIDS / III STD World
Congress, Amsterdam, NL, Congrex Holland.
Wittkowski KM (1995a). Entwicklung der HIV-Epidemie bei Hämophilen anhand der Zahlen des Robert-
Koch-Instituts (Berlin). 25 Hämophilie-Symposion, Hamburg 1994. I Scharrer and W Schramm. Berlin,
D, Springer: 20-8.
Wittkowski KM (1995b). [Epidemiology of HIV in Germany]. Gesundheitswesen 57(5): 291-8.
Wittkowski KM (1996). [Epidemiology of HIV in Germany (status: 30 June 1995)--transmission among
heterosexuals and value of HIV screening programs]. Gesundheitswesen 58(4): 229-33.
Woolhouse M, Gaunt E (2007). Ecological origins of novel human pathogens. Crit Rev Microbiol 33(4):
231-42.
Yeh SH, Wang HY, et al. (2004). Characterization of severe acute respiratory syndrome coronavirus
genomes in Taiwan: molecular epidemiology and genome evolution. Proc Natl Acad Sci U S A 101(8):
2542-7.
Zhou G, Yan G (2003). Severe acute respiratory syndrome epidemic in Asia. Emerg Infect Dis 9(12):
1608-10.
. CC-BY-ND 4.0 International licenseIt is made available under a
author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the(which was not peer-reviewed) The copyright holder for this preprint .https://doi.org/10.1101/2020.03.28.20036715doi: medRxiv preprint
... The Abstract and the main body (from Introduction to Part I of Conclusions) have been available as a prepublication on medRxiv/bioRxiv from 2020-03-31 (V1), with 2020-04-07 (V2), 2020-04-15 (V3), 2020-04-20 (V4), and 2020-04-29 (V5) updates [1], and some data were added until 2020-05-05 (ISO 8601 date format is used throughout), while several attempts for publication failed, mainly because reviewers and/or editors were concerned about two types of conclusions: 1) conclusions about the emergence of novel virus strains based on empirical data before large databases of sequencing data had been collected and 2) conclusions about the effectiveness and risks of interventions based on other epidemiological models than those that led to the lockdowns starting in late 2020-03 [2], which had been made available on the same prepublication server. To start a discussion about using epidemiological data and models earlier in an epidemic, relevant parts of this publication are still written from the early 2020 perspective with only minor changes and clarifications based on input from reviewers and editors. ...
... For most of the first three months of the epidemic, much of the response was driven by fear, stigma, or discrimination, including naming SARS-CoV-2 the "China virus" despite the fact that seasonal respiratory zoonotic pathogens typically originate in China, where live-animal markets provide chances for animal viruses to transmit to humans. The 2020-03- 23 Veterans Health Administration's COVID- 19 Response Plan envisioned a pandemic that would "last 18 months or longer and could include multiple waves of illness" (see Wittkowski [1] for reference). ...
... On 2020-02-29 (a day before the peak of the main wave or only three days after the inflection point), the Korea Disease Control and Prevention Agency (KCDC) issued a nationwide recommendation for "social distancing." Still, the recommended "social distancing" may have prevented herd immunity from developing, as suggested by the continuing low number of cases becoming infected (including some deaths) (Figure 13, bottom) (see Wittkowski [1] for reference). ...
... Further, the discussion and studies done by Wittkowski [14] and Kissler et al. [15], state that effectiveness on lockdown strategy is not known to impact the spread of the virus. Industries, such as hospitality, the airline, have taken a significant hit, followed by agriculture taking a global drop of more than 20% in demand, and manufacturing had shown a large drop in overall demand [16]. ...
Article
Full-text available
The purpose of the paper is to provide innovative emerging technology framework for community to combat epidemic situations. The paper proposes a unique outbreak response system framework based on artificial intelligence and edge computing for citizen centric services to help track and trace people eluding safety policies like mask detection and social distancing measure in public or workplace setup. The framework further provides implementation guideline in industrial setup as well for governance and contact tracing tasks. The adoption will thus lead in smart city planning and development focusing on citizen health systems contributing to improved quality of life. The conceptual framework presented is validated through quantitative data analysis via secondary data collection from researcher's public websites, GitHub repositories and renowned journals and further benchmarking were conducted for experimental results in Microsoft Azure cloud environment. The study includes selective AI-models for benchmark analysis and were assessed on performance and accuracy in edge computing environment for large scale societal setup. Overall YOLO model Outperforms in object detection task and is faster enough for mask detection and HRNetV2 outperform semantic segmentation problem applied to solve social distancing task in AI-Edge inferencing environmental setup. The paper proposes new Edge-AI algorithm for building technology-oriented solutions for detecting mask in human movement and social distance. The paper enriches the technological advancement in artificial intelligence and edge-computing applied to problems in society and healthcare systems. The framework further equips government agency, system providers to design and constructs technology-oriented models in community setup to Increase the quality of life using emerging technologies into smart urban environments.
... Important containment measures were implemented in Wuhan (China) to limit the diffusion of SARS-CoV-2 (Guan, et al., 2020;Wu & McGoogan, 2020;Zhou, Yu, Du, Fan, Liu, Xiang, et al., 2020). Despite these efforts, such new strain of coronavirus has spread all over the world (Velavan & Meyer, 2020;Wittkowski, 2020). The development and administration of vaccines is now playing a fundamental role to face the epidemic. ...
Article
While many efforts are currently devoted to vaccines development and administration, social distancing measures, including severe restrictions such as lockdowns, remain fundamental tools to contain the spread of COVID-19. A crucial point for any government is to understand, on the basis of the epidemic curve, the right temporal instant to set up a lockdown and then to remove it. Different strategies are being adopted with distinct shades of intensity. USA and Europe tend to introduce restrictions of considerable temporal length. They vary in time: a severe lockdown may be reached and then gradually relaxed. An interesting alternative is the Australian model where short and sharp responses have repeatedly tackled the virus and allowed people a return to near normalcy. After a few positive cases are detected, a lockdown is immediately set. In this paper we show that the Australian model can be generalized and given a rigorous mathematical analysis, casting strategies of the type short-term pain for collective gain in the context of sliding-mode control, an important branch of nonlinear control theory. This allows us to gain important insights regarding how to implement short-term lockdowns, obtaining a better understanding of their merits and possible limitations. Effects of vaccines administration in improving the control law’s effectiveness are also illustrated. Our model predicts the duration of the severe lockdown to be set to maintain e.g. the number of people in intensive care under a certain threshold. After tuning our strategy exploiting data collected in Italy, it turns out that COVID-19 epidemic could be e.g. controlled by alternating one or two weeks of complete lockdown with one or two months of freedom, respectively. Control strategies of this kind, where the lockdown’s duration is well circumscribed, could be important also to alleviate coronavirus impact on economy.
... A new coronavirus, causing a severe acute respiratory syndrome (COVID- 19), and being transmitted between individuals, was originally identified as 2019nCoV in Wuhan (China) in December 2019 and subsequently named SARS-CoV-2 for its 80% genome homology to that of the HCoV SARS (SARS-CoV-1) and the resemblance of its clinical manifestations to those of the aforementioned virus [1][2][3]. The epidemic was rapidly spreading from China throughout the world, to become a pandemic that, as of today, has affected more than 118 million people, causing over 2.6 million deaths [4,5]. During the first phase of massive campaign of vaccination to prevent SARS-CoV-2 infection and the related disease COVID-19, we still need to use the common measures to attempt containing pandemic such as facial masks and disinfectants, avoiding people gathering and social distancing, massive testing and tracing, or more severe restrictions like quarantine, curfew or lockdowns of all near-contact activities. ...
Article
Full-text available
(1) Background: A better understanding of COVID-19 dynamics in terms of interactions among individuals would be of paramount importance to increase the effectiveness of containment measures. Despite this, the research lacks spatiotemporal statistical and mathematical analysis based on large datasets. We describe a novel methodology to extract useful spatiotemporal information from COVID-19 pandemic data. (2) Methods: We perform specific analyses based on mathematical and statistical tools, like mathematical morphology, hierarchical clustering, parametric data modeling and non-parametric statistics. These analyses are here applied to the large dataset consisting of about 19,000 COVID-19 patients in the Veneto region (Italy) during the entire Italian national lockdown. (3) Results: We estimate the COVID-19 cumulative incidence spatial distribution, significantly reducing image noise. We identify four clusters of connected provinces based on the temporal evolution of the incidence. Surprisingly, while one cluster consists of three neighboring provinces, another one contains two provinces more than 210 km apart by highway. The survival function of the local spatial incidence values is modeled here by a tapered Pareto model, also used in other applied fields like seismology and economy in connection to networks. Model’s parameters could be relevant to describe quantitatively the epidemic. (4) Conclusion: The proposed methodology can be applied to a general situation, potentially helping to adopt strategic decisions such as the restriction of mobility and gatherings.
... respiratory infection detected in Wuhan,the largest metropolitan area in China's Hubei province which was first reported to the WHO country office in China,on dec 31,2019 [4]. Researchers around the world are working to develop potential treatments or vaccines against the respiratory diseases that have killed approximately 47,000 people infected almost a million in just a few months [5]. Coronavirus is popular as it spreads to many countries around the world. ...
Article
Background: Coronavirus disease 2019 is an infectious disease caused by severe acute respiratory syndrome. The disease was first identified on december 2019 in Wuhan,the capital of China Hubei province and has since spread globally resulting in the ongoing 2019-2020 coronavirus pandemic. Most recently,the middle east respiratory syndrome coronavirus(MERS-COV)was first identified in Saudi Arabia in 2012. In a timeline that reaches the present day which is an epidemic of cases with unexplained low respiratory infection detected in Wuhan,the largest metropolitan area in China's Hubei province which was first reported to the WHO country office in China,on dec 31,2019. Aim:To review the incidence and recovery of COVID-19 pandemic outbreak among the global population. Materials and methods: This is a literature review conducted using article sources from databases-Scopus and PubMed from september 2019 to April 2020. The articles are screened for data extraction and the characteristics of studies are tabulated. The collected data is analysed and the results are reported. Results and Conclusion: The findings of the review suggests that the recovery cases are less compared to active cases,because it is so contagious which represents the average number of people to which a single infected person transmits the virus is relatively high.
... Researchers around the world are working to develop potential treatments or vaccines against the respiratory diseases that have killed nearly 47,000 people infected almost a million in just a few months 5 . Coronavirus is popular as it spreads to many countries around the world 6 . ...
Article
Introduction: In the current COVID-19 pandemic, frontline health care workers and patients undergoing dental procedures are at the high risk of cross-infection. Most of the dental procedures require contact with the saliva, blood, and respiratory tract secretions in the oral cavity of the patient. Many patients may be carriers of the virus. It may be suggested that all patients visiting a dental office must be treated with due precautions and utmost care to prevent the cross-infection. The aim of this study was to create awareness on dental treatment during COVID-19 among the South Indian population. Materials and Methods: A self-structured questionnaire was administered to the participants to collect the data through online Google forms link. The participants were well informed about the study in detail. The data were collected and statistically analyzed. Results and Conclusion: People need to be educated about their personal hygiene which will prevent the disease spread. Awareness seminars may be conducted to educate the community about dental treatment in this pandemic period. Conclusion: Closing dental practices during the pandemic will increase the suffering of individuals in need of urgent dental care. This calls for the designing of standard guidelines for dental care during the worldwide spread of the pandemic. Awareness may be created on the use of masks, hand sanitizers, and social distancing for patients to prevent the disease spread.
Article
Full-text available
Digestive disorder symptoms in COVID-19 may be similar in form to post-infectious functional gastrointestinal disorder (PI-FGID). To cause clinical effects, SARS-CoV-2 must reach the bowels and gastric hypochlorhydria may facilitate such transit. Asian elderly are predisposed to greater infection rate and severity of COVID-19, and the high prevalence of gastric atrophy and intake of proton-pump inhibitor in this aged group might explain the risk. Persistence shedding of SARS-CoV-2 in stools indicates that faecal transmission should not be disregarded. Gut involvement in COVID-19 is mediated by angiotensin-converting enzyme 2 (ACE2) receptor, which serves as the entry point for SARS-CoV-2 in the small bowel. ACE2 dysregulation has an impact on the homeostasis of gut microbiota and altered inflammatory response. Liver injury is variable in COVID-19 and is likely a result of by-stander effects rather than actual viropathic process. Further research is needed to understand if gut involvement is a cause or effect of SARS-CoV-2.
Article
Full-text available
Understanding the SARS-CoV-2 dynamics has been subject of intense research in the last months. In particular, accurate modeling of lockdown effects on human behaviour and epidemic evolution is a key issue in order e.g. to inform health-care decisions on emergency management. In this regard, the compartmental and spatial models so far proposed use parametric descriptions of the contact rate, often assuming a time-invariant effect of the lockdown. In this paper we show that these assumptions may lead to erroneous evaluations on the ongoing pandemic. Thus, we develop a new class of nonparametric compartmental models able to describe how the impact of the lockdown varies in time. Our estimation strategy does not require significant Bayes prior information and exploits regularization theory. Hospitalized data are mapped into an infinite-dimensional space, hence obtaining a function which takes into account also how social distancing measures and people’s growing awareness of infection’s risk evolves as time progresses. This also permits to reconstruct a continuous-time profile of SARS-CoV-2 reproduction number with a resolution never reached before in the literature. When applied to data collected in Lombardy, the most affected Italian region, our model illustrates how people behaviour changed during the restrictions and its importance to contain the epidemic. Results also indicate that, at the end of the lockdown, around $$12\%$$ 12 % of people in Lombardy and $$5\%$$ 5 % in Italy was affected by SARS-CoV-2, with the fatality rate being 1.14%. Then, we discuss how the situation evolved after the end of the lockdown showing that the reproduction number dangerously increased in the summer, due to holiday relax, reaching values larger than one on August 1, 2020. Finally, we also document how Italy faced the second wave of infection in the last part of 2020. Since several countries still observe a growing epidemic and others could be subject to other waves, the proposed reproduction number tracking methodology can be of great help to health care authorities to prevent SARS-CoV-2 diffusion or to assess the impact of lockdown restrictions on human behaviour to contain the spread.
Article
Epidemics caused by airborne viruses in cities with large populations create a big problem as in the current COVID-19 pandemic. Cramped lifestyle, busy workplaces, crowded public transportation, and higher household member counts are responsible for the transmission of the disease. In Turkey, Istanbul has taken the lead in the number of cases since the beginning of the epidemic. The excess population density is the major cause for disease transmission. It is essential to monitor the contaminated regions with geographical information systems on city maps. Outbreak maps visualize and help analyze the patterns of transmission and serve as a communication and education tool. A dynamic heat map video of SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) polymerase chain reaction positive cases in a county of Istanbul was generated. The heat map visualizes how the epidemic spread to all the districts and the cumulative cases increased in one county of Istanbul with real attack rates.
Chapter
The chapter reviews primary-source virological and epidemiological studies to profile the COVID-19 virus. Key epidemiological concepts are introduced and various methods of mitigating viruses are discussed. The social nature of virus communicability and the roles of interpersonal distance and high-contact cultures as media for the transmission of the virus are detailed.
Article
Full-text available
The beginning of 2020 has seen the emergence of COVID-19 outbreak caused by a novel coronavirus, Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2). There is an imminent need to better understand this new virus and to develop ways to control its spread. In this study, we sought to gain insights for vaccine design against SARS-CoV-2 by considering the high genetic similarity between SARS-CoV-2 and SARS-CoV, which caused the outbreak in 2003, and leveraging existing immunological studies of SARS-CoV. By screening the experimentally-determined SARS-CoV-derived B cell and T cell epitopes in the immunogenic structural proteins of SARS-CoV, we identified a set of B cell and T cell epitopes derived from the spike (S) and nucleocapsid (N) proteins that map identically to SARS-CoV-2 proteins. As no mutation has been observed in these identified epitopes among the 120 available SARS-CoV-2 sequences (as of 21 February 2020), immune targeting of these epitopes may potentially offer protection against this novel virus. For the T cell epitopes, we performed a population coverage analysis of the associated MHC alleles and proposed a set of epitopes that is estimated to provide broad coverage globally, as well as in China. Our findings provide a screened set of epitopes that can help guide experimental efforts towards the development of vaccines against SARS-CoV-2.
Article
Full-text available
An outbreak of coronavirus disease 2019 (COVID-19) caused by the 2019 novel coronavirus (SARS-CoV-2) began in Wuhan, Hubei Province, China in December 2019, and has spread throughout China and to 31 other countries and territories, including the United States (1). As of February 23, 2020, there were 76,936 reported cases in mainland China and 1,875 cases in locations outside mainland China (1). There have been 2,462 associated deaths worldwide; no deaths have been reported in the United States. Fourteen cases have been diagnosed in the United States, and an additional 39 cases have occurred among repatriated persons from high-risk settings, for a current total of 53 cases within the United States. This report summarizes the aggressive measures (2,3) that CDC, state and local health departments, multiple other federal agencies, and other partners are implementing to slow and try to contain transmission of COVID-19 in the United States. These measures require the identification of cases and contacts of persons with COVID-19 in the United States and the recommended assessment, monitoring, and care of travelers arriving from areas with substantial COVID-19 transmission. Although these measures might not prevent widespread transmission of the virus in the United States, they are being implemented to 1) slow the spread of illness; 2) provide time to better prepare state and local health departments, health care systems, businesses, educational organizations, and the general public in the event that widespread transmission occurs; and 3) better characterize COVID-19 to guide public health recommendations and the development and deployment of medical countermeasures, including diagnostics, therapeutics, and vaccines. U.S. public health authorities are monitoring the situation closely, and CDC is coordinating efforts with the World Health Organization (WHO) and other global partners. Interim guidance is available at https://www.cdc.gov/coronavirus/index.html. As more is learned about this novel virus and this outbreak, CDC will rapidly incorporate new knowledge into guidance for action by CDC, state and local health departments, health care providers, and communities.
Article
Full-text available
The global spread of the 2019-nCoV is continuing and is fast moving, as indicated by the WHO raising the risk assessment to high. In this article, we provide a preliminary phylodynamic and phylogeographic analysis of this new virus. A Maximum Clade Credibility tree has been built using the 29 available whole genome sequences of 2019-nCoV and two whole genome sequences that are highly similar sequences from Bat SARS-like Coronavirus available in GeneBank. We are able to clarify the mechanism of transmission among the countries which have provided the 2019-nCoV sequence isolates from their patients. The Bayesian phylogeographic reconstruction shows that the 2019–2020 nCoV most probably originated from the Bat SARS-like Coronavirus circulating in the Rhinolophus bat family. In agreement with epidemiological observations, the most likely geographic origin of the new outbreak was the city of Wuhan, China, where 2019-nCoV time of the most recent common ancestor emerged, according to molecular clock analysis, around November 25th, 2019. These results, together with previously recorded epidemics, suggest a recurring pattern of periodical epizootic outbreaks due to Betacoronavirus. Moreover, our study describes the same population genetic dynamic underlying the SARS 2003 epidemic, and suggests the urgent need for the development of effective molecular surveillance strategies of Betacoronavirus among animals and Rhinolophus of the bat family.
Article
Full-text available
Coronaviruses are the well-known cause of severe respiratory, enteric and systemic infections in a wide range of hosts including mammals, fish, and avian. The scientific interest on coronaviruses increased after the emergence of Severe Acute Respiratory Syndrome coronavirus (SARS-CoV) outbreaks in 2002-2003 followed by Middle East Respiratory Syndrome CoV (MERS-CoV). This decade’s first CoV, named 2019-nCoV, emerged from Wuhan, China, and declared as “Public Health Emergency of International Concern” on January 30th, 2020 by the World Health Organization (WHO). As on February 4, 2020, 425 deaths reported in China only and one death outside China (Philippines). In a short span of time, the virus spread has been noted in 24 countries. The zoonotic transmission (animal-to-human) is suspected as the route of disease origin. The genetic analyses predict bats as the most probable source of 2019-nCoV though further investigations needed to confirm the origin of the novel virus. The ongoing nCoV outbreak highlights the hidden wild animal reservoir of the deadly viruses and possible threat of spillover zoonoses as well. The successful virus isolation attempts have made doors open for developing better diagnostics and effective vaccines helping in combating the spread of the virus to newer areas.
Article
Full-text available
Background: Human immunodeficiency virus (HIV) is a global public health issue, with an estimated 36.9 million people living with HIV in 2017. HIV has been reportable in Canada since 1985 and the Public Health Agency of Canada (PHAC) continues to monitor trends in new HIV diagnoses. Objective: The objective of this surveillance report is to provide an overview of the epidemiology of all reported diagnoses of HIV in Canada since 1985 with a focus on 2018 overall, and by geographic location, age group, sex, and exposure category. Methods: PHAC monitors HIV through the national HIV/AIDS Surveillance System, a passive, case-based system that collates nonnominal data that is voluntarily submitted by all Canadian provinces and territories. Descriptive epidemiological analyses were conducted on national data and those relating to specific populations provided by Immigration, Refugees and Citizenship Canada and the Canadian Perinatal HIV Surveillance Program. Results: In 2018, a total of 2,561 HIV diagnoses were reported in Canada, an increase of 8.2% compared with 2017. The national diagnosis rate increased to 6.9 per 100,000 population in 2018 from 6.5 per 100,000 population in 2017. Saskatchewan reported the highest provincial diagnosis rate at 14.9 per 100,000 population. The 30-39 year age group continued to have the highest HIV diagnosis rate at 15.4 per 100,000 population. Overall, the diagnosis rate for males continued to be higher than that of females (9.8 versus 4.0 per 100,000 population, respectively); however, females experienced a larger increase in reported cases and diagnosis rate. The gay, bisexual and other men who have sex with men (gbMSM) exposure category continued to represent the highest proportion of all reported adult cases (41.4%), though the proportion has decreased over time. Five perinatal HIV transmissions were documented, three were related to the mother not receiving perinatal antiretroviral therapy prophylaxis. Conclusion: The number and rate of reported HIV cases in Canada increased in 2018, gbMSM continued to account for the largest exposure category and the number and rate of reported HIV cases among women increased. PHAC will continue to work with its national partners to refine the collection, analysis and publication of national data to better understand the burden of HIV in Canada.
Article
Full-text available
This article provides an overview of the severe acute respiratory syndrome (SARS) epidemics in mainland China and of what we have learned since the outbreak. The epidemics spanned a large geographical extent but clustered in two regions: first in Guangdong Province, and about 3 months later in Beijing and its surrounding areas. The resulting case fatality ratio of 6.4% was less than half of that in other SARS-affected countries and regions, partly due to younger-aged patients and a higher proportion of community-acquired infections. Strong political commitment and a centrally coordinated response were most important for controlling SARS. The long-term economic consequence of the epidemic was limited. Many recovered patients suffered from avascular osteonecrosis, as a consequence of corticosteroid usage during their infection. The SARS epidemic provided valuable experience and lessons relevant in controlling outbreaks of emerging infectious diseases, and has led to fundamental reforms of the Chinese health system. Additionally, the epidemic has substantially improved infrastructures, surveillance systems, and capacity to response to health emergencies. In particular, a comprehensive nationwide internet-based disease reporting system was established.
Article
Mutations can reveal connections between outbreaks—but it's easy to overinterpret them
Article
Background: A novel human coronavirus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), was identified in China in December 2019. There is limited support for many of its key epidemiologic features, including the incubation period for clinical disease (coronavirus disease 2019 [COVID-19]), which has important implications for surveillance and control activities. Objective: To estimate the length of the incubation period of COVID-19 and describe its public health implications. Design: Pooled analysis of confirmed COVID-19 cases reported between 4 January 2020 and 24 February 2020. Setting: News reports and press releases from 50 provinces, regions, and countries outside Wuhan, Hubei province, China. Participants: Persons with confirmed SARS-CoV-2 infection outside Hubei province, China. Measurements: Patient demographic characteristics and dates and times of possible exposure, symptom onset, fever onset, and hospitalization. Results: There were 181 confirmed cases with identifiable exposure and symptom onset windows to estimate the incubation period of COVID-19. The median incubation period was estimated to be 5.1 days (95% CI, 4.5 to 5.8 days), and 97.5% of those who develop symptoms will do so within 11.5 days (CI, 8.2 to 15.6 days) of infection. These estimates imply that, under conservative assumptions, 101 out of every 10 000 cases (99th percentile, 482) will develop symptoms after 14 days of active monitoring or quarantine. Limitation: Publicly reported cases may overrepresent severe cases, the incubation period for which may differ from that of mild cases. Conclusion: This work provides additional evidence for a median incubation period for COVID-19 of approximately 5 days, similar to SARS. Our results support current proposals for the length of quarantine or active monitoring of persons potentially exposed to SARS-CoV-2, although longer monitoring periods might be justified in extreme cases. Primary funding source: U.S. Centers for Disease Control and Prevention, National Institute of Allergy and Infectious Diseases, National Institute of General Medical Sciences, and Alexander von Humboldt Foundation.
Article
The current corona virus disease 2019 outbreak caused by severe acute respiratory syndrome coronavirus 2 started in Wuhan, China in December 2019 and has put the world on alert. To safeguard Chinese citizens and to strengthen global health security, China has made great efforts to control the epidemic. Many in the global community have joined China to limit the epidemic. However, discrimination and prejudice driven by fear or misinformation have been flowing globally, superseding evidence and jeopardizing the anti-severe acute respiratory syndrome coronavirus 2 efforts. We analyze this phenomenon and its underlying causes and suggest practical solutions.
Article
Background: In late December, 2019, patients presenting with viral pneumonia due to an unidentified microbial agent were reported in Wuhan, China. A novel coronavirus was subsequently identified as the causative pathogen, provisionally named 2019 novel coronavirus (2019-nCoV). As of Jan 26, 2020, more than 2000 cases of 2019-nCoV infection have been confirmed, most of which involved people living in or visiting Wuhan, and human-to-human transmission has been confirmed. Methods: We did next-generation sequencing of samples from bronchoalveolar lavage fluid and cultured isolates from nine inpatients, eight of whom had visited the Huanan seafood market in Wuhan. Complete and partial 2019-nCoV genome sequences were obtained from these individuals. Viral contigs were connected using Sanger sequencing to obtain the full-length genomes, with the terminal regions determined by rapid amplification of cDNA ends. Phylogenetic analysis of these 2019-nCoV genomes and those of other coronaviruses was used to determine the evolutionary history of the virus and help infer its likely origin. Homology modelling was done to explore the likely receptor-binding properties of the virus. Findings: The ten genome sequences of 2019-nCoV obtained from the nine patients were extremely similar, exhibiting more than 99·98% sequence identity. Notably, 2019-nCoV was closely related (with 88% identity) to two bat-derived severe acute respiratory syndrome (SARS)-like coronaviruses, bat-SL-CoVZC45 and bat-SL-CoVZXC21, collected in 2018 in Zhoushan, eastern China, but were more distant from SARS-CoV (about 79%) and MERS-CoV (about 50%). Phylogenetic analysis revealed that 2019-nCoV fell within the subgenus Sarbecovirus of the genus Betacoronavirus, with a relatively long branch length to its closest relatives bat-SL-CoVZC45 and bat-SL-CoVZXC21, and was genetically distinct from SARS-CoV. Notably, homology modelling revealed that 2019-nCoV had a similar receptor-binding domain structure to that of SARS-CoV, despite amino acid variation at some key residues. Interpretation: 2019-nCoV is sufficiently divergent from SARS-CoV to be considered a new human-infecting betacoronavirus. Although our phylogenetic analysis suggests that bats might be the original host of this virus, an animal sold at the seafood market in Wuhan might represent an intermediate host facilitating the emergence of the virus in humans. Importantly, structural analysis suggests that 2019-nCoV might be able to bind to the angiotensin-converting enzyme 2 receptor in humans. The future evolution, adaptation, and spread of this virus warrant urgent investigation. Funding: National Key Research and Development Program of China, National Major Project for Control and Prevention of Infectious Disease in China, Chinese Academy of Sciences, Shandong First Medical University.