Cases

Main articles: COVID-19 pandemic by country and territory and COVID-19 pandemic

cases
Official case counts refer to the number of people who have been tested for COVID-19 and
whose test has been confirmed positive according to official protocols. [57][58] Many countries,
early on, had official policies to not test those with only mild symptoms. [59][60] An analysis of
the early phase of the outbreak up to 23 January estimated 86 percent of COVID-19
infections had not been detected, and that these undocumented infections were the source
for 79 percent of documented cases.[61] Several other studies, using a variety of methods,
have estimated that numbers of infections in many countries are likely to be considerably
greater than the reported cases.[62][63]
On 9 April 2020, preliminary results found that 15 percent of people tested in Gangelt, the
centre of a major infection cluster in Germany, tested positive for antibodies. [64] Screening
for COVID-19 in pregnant women in New York City, and blood donors in the Netherlands,
has also found rates of positive antibody tests that may indicate more infections than
reported.[65][66] Seroprevalence based estimates are conservative as some studies shown
that persons with mild symptoms do not have detectable antibodies. [67] Some results (such
as the Gangelt study) have received substantial press coverage without first passing
through peer review.[68]
Analysis by age in China indicates that a relatively low proportion of cases occur in
individuals under 20.[69] It is not clear whether this is because young people are less likely to
be infected, or less likely to develop serious symptoms and seek medical attention and be
tested.[70] A retrospective cohort study in China found that children were as likely to be
infected as adults.[71]
Initial estimates of the basic reproduction number (R0) for COVID-19 in January were
between 1.4 and 2.5,[72] but a subsequent analysis concluded that it may be about 5.7 (with
a 95 percent confidence interval of 3.8 to 8.9).[73] R0 can vary across populations and is not
to be confused with the effective reproduction number (commonly just called R), which
takes into account effects such as social distancing and herd immunity. By mid-May 2020,
the effective R was close to or below 1.0 in many countries, meaning the spread of the
disease in these areas at that time was stable or decreasing. [74]

Epidemic curve of COVID-19 by date of report

 
 

Semi-log plot of daily new cases of COVID-19 (seven-day average) in the world and top five
current countries (mean with deaths)
 

Daily confirmed cases by country per million people, plotted on a logarithmic scale
 
 

COVID-19 total cases per 100 000 population from selected countries [75]
 

COVID-19 active cases per 100 000 population from selected countries [75]

Deaths
Main articles: COVID-19 pandemic deaths and COVID-19 pandemic death rates by
country
Further information: List of deaths due to COVID-19

Deceased in a 16 m (53 ft) "mobile morgue" outside a hospital in Hackensack, New Jersey

Most people who contract COVID-19 recover. For those who do not, the time between the
onset of symptoms and death usually ranges from 6 to 41 days, typically about 14 days.
  As of 26 August 2020, approximately 821,000[5] deaths had been attributed to COVID-19.
[76]

In China, as of 14 June, about 80 percent of deaths were recorded in those aged over 60,
and 75 percent had pre-existing health conditions including cardiovascular
diseases and diabetes.[77] Individuals of any age with COPD, obesity, type 2 diabetes
mellitus and other underlying health conditions are at increased risk of severe illness from
COVID-19.[78][79]
The first confirmed death was in Wuhan on 9 January 2020. [80] The first death outside of
China occurred on 1 February in the Philippines,[81] and the first death outside Asia was in
France on 14 February.[82]
Official deaths from COVID-19 generally refer to people who died after testing positive
according to protocols. This may ignore deaths of people who die without having been
tested.[83] Conversely, deaths of people who had underlying conditions may lead to over-
counting.[84] Comparison of statistics for deaths for all causes versus the seasonal average
indicates excess mortality in many countries.[85][86] In the worst affected areas, mortality has
been several times higher than average. In New York City, deaths have been four times
higher than average, in Paris twice as high, and in many European countries, deaths have
been on average 20 to 30 percent higher than normal. [85] This excess mortality may include
deaths due to strained healthcare systems and bans on elective surgery.[87]
Multiple measures are used to quantify mortality.[88] These numbers vary by region and over
time, influenced by testing volume, healthcare system quality, treatment options,
government response,[89][90][91] time since the initial outbreak, and population characteristics,
such as age, sex, and overall health. [92] Some countries (like Belgium) include deaths from
suspected cases of COVID-19, regardless of whether the person was tested, resulting in
higher numbers compared to countries that include only test-confirmed cases. [93]
The death-to-case ratio reflects the number of deaths attributed to COVID-19 divided by the
number of diagnosed cases within a given time interval. Based on Johns Hopkins
University statistics, the global death-to-case ratio is 3.4 percent (821,909 deaths for
24,011,502 cases) as of 26 August 2020. [5] The number varies by region.[94]
Other measures include the case fatality rate (CFR), which reflects the percentage of
diagnosed people who die from a disease, and the infection fatality rate (IFR), which
reflects the percentage of infected (diagnosed and undiagnosed) who die from a disease.
These statistics are not timebound and follow a specific population from infection through
case resolution. Our World in Data states that as of 25 March 2020 the IFR cannot be
accurately calculated as neither the total number of cases nor the total deaths, is known.
[92]
 In February the Institute for Disease Modeling estimated the IFR as 0.94 percent (95-
percent confidence interval 0.37–2.9), based on data from China. [95][96] The University of
Oxford's Centre for Evidence-Based Medicine (CEBM) estimated a global CFR of 0.8 to 9.6
percent (last revised 30 April) and IFR of 0.10 percent to 0.41 percent (last revised 2 May),
acknowledging that this will vary between populations due to differences in demographics.
[97]
 The CDC estimates for planning purposes that the fatality rate among those who are
symptomat