Our work belongs to everyone
- All our code is open-source
- All our research and visualizations are free for everyone to use for all purposes
- The positive rate: A crucial metric for understanding the pandemic
- The scale of testing compared to the scale of the outbreak
- How many tests are performed each day?
- Tests per confirmed case
- World map: total tests performed relative to the size of population
- Testing vs GDP per capita
- Testing and contact tracing policy
No country knows the total number of people infected with COVID-19. All we know is the infection status of those who have been tested. All those who have a lab-confirmed infection are counted as confirmed cases.
This means that the counts of confirmed cases depend on how much a country actually tests. Without testing there is no data.
Testing is our window onto the pandemic and how it is spreading. Without data on who is infected by the virus we have no way of understanding the pandemic. Without this data we can not know which countries are doing well, and which are just underreporting cases and deaths.
To interpret any data on confirmed cases we need to know how much testing for COVID-19 the country actually does.
Because testing is so very crucial to understanding the spread of the pandemic and responding appropriately we have focused our efforts on building a global dataset on COVID-19 testing.
- The testing dataset is updated around twice a week. The latest version is always available in the section below.
- And as with all our work, it is freely accessible for everyone. The data can be downloaded here on GitHub.
The positive rate: A crucial metric for understanding the pandemic
Here we show the share of tests returning a positive result – known as the positive rate.
This metric offers us two key insights: firstly as a measure of how adequately countries are testing; and secondly to help us understand the spread of the virus, in conjunction with data on confirmed cases.
The positive rate is a good metric for how adequately countries are testing because it indicates the level of testing relative to the size of the outbreak. To be able to properly monitor and control the spread of the virus, countries with more widespread outbreaks need to do more testing.
We see enormous differences across countries:
- Some countries, like Australia, South Korea and Uruguay have a positive rate of less than 1% – it takes hundreds, or even thousands of tests to find one case in these countries.
- Others, such as Mexico and Bolivia, have positive rates of 20%–50% or even more. In these countries a case is found for every few tests conducted.
According to criteria published by WHO in May, a positive rate of less than 5% is one indicator that the epidemic is under control in a country.1
Because limited testing makes it likely that many cases will be missed, the positive rate can also help our understanding of the spread of the virus. In countries with a high positive rate, the number of confirmed cases is likely to represent only a small fraction of the true number of infections. And where the positive rate is rising in a country, this can suggest the virus is actually spreading faster than the growth seen in confirmed cases.
The scale of testing compared to the scale of the outbreak
This scatter chart provides another way of seeing the extent of testing relative to the scale of the outbreak in different countries.
The chart shows the daily number of tests (vertical axis) against the daily number of new confirmed cases (horizontal axis), per million people.
Looking down the chart, we see some countries doing ten or hundred times fewer tests than other countries with a similar number of new confirmed cases.
Conversely, looking to the right, we see some countries find ten or a hundred times more cases than others out a similar number of tests.
Where the number of confirmed cases is high relative to the extent of testing, this suggests that there may not be enough tests being carried out to properly monitor the outbreak. In such countries, the true number of infections may be far higher than the number of confirmed cases.
How many tests are performed each day?
Related charts:
This chart shows the number of daily tests per thousand people. Because the number of tests is often volatile from day to day, we show the figures as a seven-day rolling average.
What is counted as a test?
The number of tests does not refer to the same in each country – one difference is that some countries report the number of people tested, while others report the number of tests (which can be higher if the same person is tested more than once). And other countries report their testing data in a way that leaves it unclear what the test count refers to exactly.
We indicate the differences in the chart and explain them in detail in our accompanying source descriptions.
How to interact with this chart
- By clicking on Add country you can show and compare the data for any country in the world you are interested in.
- If you move both ends of the time-slider to a single point you will see a bar chart for this point in time.
- You can switch to the ‘MAP’ tab.
In all our charts you can download the data
We want everyone to built on top of our work and therefore we always make all our data always available for download. Click on the ‘Data’-tab below the chart and you can download the shown data for all countries in a simple to use csv file.
Tests per confirmed case
Another way of looking at the extent of testing relative to the scale of the outbreak is to ask: How many tests does a country do to find one COVID-19 case?
It is simply the inverse of our data on the positive rate.
Countries that do very few tests per confirmed case are unlikely to be testing widely enough to find all cases. The WHO has suggested around 10 – 30 tests per confirmed case as a general benchmark of adequate testing.2
World map: total tests performed relative to the size of population
Related charts:
This map shows you how the total number of tests per thousand people compares across all countries in our dataset.
How to interact with this chart
- By moving the time slider (below the map) you can see how testing coverage has changed over time.
- You can focus on a particular world region using the dropdown menu to the top-right of the map.
- Hovering over a country lets you see the exact number.
Testing vs GDP per capita
This chart plots the total number of tests performed per thousand people against GDP per capita. We see that richer countries in general have done more testing.
The differences in the extent of testing between rich and poor countries is large. While poorer countries such as Cote d’Ivoire and Nigeria have done only a few tests per thousand people, richer countries like Australia, Denmark and the United Arab Emirates have done hundreds of tests per thousand people.
Testing and contact tracing policy
These two charts map testing and contact tracing policies over time, using data from the Coronavirus Government Response Tracker (OxCGRT) produced by researchers at the Blavatnik School of Government, University of Oxford.
The importance of testing
Testing is our window onto the pandemic and how it is spreading. Without testing we have no way of understanding the pandemic.
It is one of our most important tools in the fight to slow and reduce the spread and impact of the virus. Tests allow us to identify infected individuals, guiding the medical treatment that they receive. It enables the isolation of those infected and the tracing and quarantining of their contacts.3 And it can help allocate medical resources and staff more efficiently.4
In addition, testing for COVID-19 also informs our understanding of the pandemic and the risks it poses in different populations.
This knowledge is important if we are to properly assess the interventions that should be implemented, including very costly interventions such as social distancing and the shutdown of entire regions and industries.
Why data on testing is needed
Without data on COVID-19 we cannot possibly understand how the pandemic is progressing.
Without data we cannot respond appropriately to the threat; neither as individuals nor as a society. Nor can we learn where countermeasures against the pandemic are working.
The number of confirmed cases is what informs us about the development of the pandemic.
But the confirmation of a case is based on a test. The World Health Organization defines a confirmed case as “a person with laboratory confirmation of COVID-19 infection”.5
Reliable data on testing is therefore necessary to assess the reliability of the data that informs us about the spread of the pandemic: the data on cases and deaths.
Different types of tests for COVID-19
There are many different technologies for COVID-19 testing, some currently available and some still in development. Trackers of the development, regulatory status and commercial release of different types of COVID-19 test are being compiled by Johns Hopkins University and the medical industry news website, 360Dx.
Broadly, we can divide these different tests into two kinds:
- those that test for the presence of the virus, aiming to establish whether an individual is currently infected. The most common way of performing a test of the first type is with a ‘PCR’ test.6
- those that test for the presence of antibodies, aiming to establish whether an individual has been infected at some point in the past.
Currently, we aim to include only these PCR tests in our testing dataset.
We do this for the following reasons:
- Our focus is on using testing data to help properly interpret the data we have on confirmed cases and deaths. Case confirmation is generally based on a positive result from a PCR test, in line with WHO recommendations.7 So including antibody tests in our figures would mean they were less useful for this purpose.
- Other kinds of test beyond PCR are not yet being widely used. This also means that data on how many of these tests have been conducted is very limited. Including such data in our counts when it is available would reduce the comparability of our data across countries.
- There are technical differences in how results from these different tests should be interpreted. Current data suggests that other existing testing technologies are subject to very different rates of false positive and false negative results than PCR tests.8 This is another reason why aggregating the data across these different types of test is not the best way of using testing data to help us understand the epidemic.
Our checklist for COVID-19 testing data
Different countries publish their testing data according different definitions. In order to make meaningful comparisons between countries and over time, the figures need to be interpreted alongside an understanding of these differences.
This means that, in addition to the numbers, detailed descriptions are needed to make clear what the numbers precisely mean. For each country in our dataset, we provide source descriptions detailing all the information that we have been able to find. However, many countries currently do not provide sufficient documentation.
For citizens to trust and understand the published data, and for countries to learn from each other, it is crucial that every country provides the data on testing in a clearly documented way.
To help guide users and producers of testing data, we provide the following checklist of questions. Clear answers to these questions are what is needed to properly interpret and compare published numbers.
1) Is there no data – or it is just hard to find?
Many countries are not yet providing official figures. Others do not do so on a regular basis. The first question to ask, then, is if there is any testing data for a given country.
Equally important is to make the available data findable. Currently, the available data is often not easy to find, because some countries are releasing figures at unpredictable intervals in ad-hoc locations (including social media or press conferences).
2) What testing technologies are being used?
There are many different technologies for COVID-19 testing, some of which are already implemented, some currently available but not yet rolled out, and some still in development. As we discuss here, these different tests are used with different objectives in mind, and there are technical differences in how results from these different testing technologies should be interpreted.
It’s critical that governments provide a detailed and explicit account of the technologies that are being implemented as they get rolled out, disaggregating the test results accordingly. For citizens to trust and understand the published data, and for epidemiologists to incorporate the data into the models that inform public policy, it is crucial that every country provides the data on testing in a clearly documented way.
3) Do numbers refer to ‘performed tests’ or ‘individuals tested’?
The number of tests performed is different to the number of individuals tested. The reason for this is that it is common for COVID-19 testing that the same person is tested more than once.
Some countries report tests performed, while others report the number of individuals tested.
The source description should state clearly what is counted.
4) Are negative results included? Are pending results included?
It needs to be clear whether or not figures for the total number of tests performed, or the number of people tested, include negative test results, as well as the number of tests that are pending results.
Many sources report the number of individuals who are ‘suspected’ or have been ‘ruled out’. To be reliably included in test counts, it needs to be explicit whether such categories reflect the number of people who are awaiting test results or have tested negatively.
5) Do the figures include all tests conducted in the country, or only some?
Figures reported by countries may only be partial if not all laboratories are reporting to the central authority.
The scope of testing data should be made explicit by the source. For instance, the US CDC make it clear that their figures do not include tests conducted in private labs.
6) Are all regions and laboratories within a country submitting data on the same basis?
Answers to the questions above may vary from region to region. In order to assess the reliability of aggregate testing data, it needs to be clear if heterogenous data is being summed together.
The US COVID Tracking Project, for instance makes it clear that their US totals combine data for tests performed and individuals tested, depending on which is reported by individual states.
7) What period do the published figures refer to?
Cumulative counts of the total number of tests should make clear the date from which the count begins. The key question that needs to be answered is whether the figures published at some date (attempt to) include all tests conducted up to that date.
Because the reporting of tests can take several days, for some countries figures for the last few days may not yet be complete. It needs to be made clear by the source if this may be the case. The US CDC, for instance, makes this clear.
8) Are there any issues that affect the comparability of the data over time?
If we want to look at how testing figures are changing over time, we need to know how any of the factors discussed above may have changed too.
The Netherlands, for instance, makes it clear that not all labs were included in national estimates from the start. As new labs get included, their past cumulative total gets added to the day they begin reporting, creating spikes in the time series.
9) What are the typical testing practices in the country?
Having a sense of how often and when individuals are tested, can help the users of these statistics understand how estimates of tests performed and individuals tested might relate to each other.
For instance, how many tests does a case investigation require? What are the eligibility criteria to be tested? Are health workers, or other specific groups, being routinely retested?
10) Might any of the information above be lost in translation?
People accessing data published in a language in which they are not fluent may misinterpret the data by mistranslating the provided text, which often includes technical terms.
Many countries report testing data in multiple languages – this helps disseminate the information to a broader audience, whilst helping prevent misinterpretations.
Download the data
Last update: August 24, 2020 21:00, London time.
(Our testing dataset is updated around twice a week.)
We make our full testing dataset, alongside detailed source descriptions, available on GitHub here. Our testing dataset is updated around twice a week.
About the dataset
Our goal at Our World in Data is to provide testing data over time for many countries around the world.
To do this, we collect publicly available information published by official sources on a regular basis. Our dataset is updated around twice a week.
Alongside the data, we also aim to provide a good understanding of the definitions used and any important limitations they might have. Our checklist of questions about testing data is what guides our efforts.
We present this information in source descriptions for each country included in the dataset. But in many cases sources do not yet provide the detailed descriptions of the data we would like.
Which countries do we have testing data for?
In this map countries for which testing data can be found in our dataset are shown in blue.
As you can see, we do not have data for all countries in our dataset. Those countries where we have looked, but have not been able to find any official sources of testing data are shown in red. Countries which we are either in the process of adding to our dataset, or for which we have not yet been able to look for data are shown in grey.
Countries publish their testing data at different frequencies: some provide daily updates, others only weekly, and some only publish figures on an ad-hoc basis at longer intervals.
Because of this, the most recent data we have for different countries refers to different dates. This chart shows you how up to date the latest data is for each country in our dataset.
Our dataset covers 69% of the world’s population:
- 30% is covered with figures relating to 23 August 2020 or later;
- 63% is covered with figures relating to 17 August 2020 or later.
Source information country-by-country
This is the list of the 94 countries for which we have data.
At the very bottom of this post, you can also find the list of countries for which we have attempted to collect data but could not find official sources.
You can download the full dataset alongside the detailed source descriptions on GitHub.
Or else click to jump to the detailed source description and the latest data for that country:
→ View the country profile of Argentina for the coronavirus pandemic.
Source #1: Government of Argentina
Short description: The number of people tested.
Latest estimate: 0.25 daily tests per thousand people (as of 20 August 2020).
Detailed description:
The government of Argentina publishes a complete time series of people tested on its open data portal: “Information updated daily at 8:00 p.m. notifying COVID-19 CASES registered in the country with a cut of the day at 5:45 p.m.”.
The dataset contains one row per person. We only keep rows with a non-empty diagnosis date (“fecha_diagnostico”), thereby filtering out people who were not tested. We then create a daily time series by calculating the number of rows for each date.
Source #2: Government of Argentina
Short description: The number of tests performed.
Latest estimate: 0.3 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The government of Argentina publishes a complete time series of tests performed on its open data portal: “Information updated daily at 8:00 p.m. notifying COVID-19 [tests] registered in the country at 17:45.”
The dataset is split by geographical location. We therefore create a daily time series by calculating the total number of tests (“total”) for each date (“fecha”).
It is unclear whether the reported figures include pending test results.
→ View the country profile of Australia for the coronavirus pandemic.
Source: Australian Government Department of Health
Short description: The number of tests performed.
Latest estimate: 2.29 daily tests per thousand people (as of 23 August 2020).
Detailed description:
We construct a time series of the cumulative number of tests performed to date using data provided by the Australian Government Department of Health through daily health alerts and weekly epidemiological reports.
The weekly epidemiological reports make it clear that the figures relate to diagnostic testing. It is unclear whether the reported figures include pending test results. It is also unclear when the first test was conducted, since we have not been able to find official data prior to 22 March 2020 (at which point 143,056 tests had been performed).
The daily health alerts have provided testing figures since 5 April 2020, whereas the weekly epidemiological reports have provided testing figures since 22 March 2020. In the daily health alerts, testing figures are reported within the “Coronavirus (COVID-19) at a glance” infographic. View the entire collection of these infographics dating back to 5 April 2020.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Austria for the coronavirus pandemic.
Source: Austria Health Ministry
Short description: The number of tests performed
Latest estimate: 1.22 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Ministry for Health publishes daily updates of the COVID-19 situation at www.sozialministerium.at/Informationen-zum-Coronavirus/Neuartiges-Coronavirus-(2019-nCov).html, which include data on the cumulative number of tests performed to date. We construct a daily time series using Web Archive snapshots of these updates.
It is unclear whether the reported figures include pending test results.
A footnote explains that “The number ‘testing’ is the total number of all Covid tests carried out. The data for this are largely reported by the performing laboratories.”
The same figures are also provided by the Ministry for Health in this official dashboard.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Bahrain for the coronavirus pandemic.
Source: Bahrain Ministry of Health
Short description: The source reports the ‘number of assessments’ conducted. It is unclear whether this refers to the total number of tests conducted, or the number of people tested.
Latest estimate: 5.67 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Bahrainian Ministry of Health publishes frequent updates on the number of confirmed cases, status of existing cases and number of assessments conducted.
These figures represent the cumulative tests to that given date. It is not clear whether this refers to the total number of tests conducted, or the number of people tested.
Using web archives we can construct a time-series of tests conducted over time based on these frequent updates. It is not clear when testing first began; data is only available from 5 March 2020 where it was reported that 5,334 tests had been conducted.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Bangladesh for the coronavirus pandemic.
Source: Government of Bangladesh
Short description: The number of samples tested.
Latest estimate: 0.08 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The Government of Bangladesh provides data on the total number of samples tested (“মোট পরিক্ষাকৃত নমুনার সংখ্যা”) to date in this official dashboard. The full time series of samples tested can be downloaded in the detailed dashboard, which dates back to 4 March 2020.
It is not clear whether the reported figures include samples in which the test results are pending.
→ View the country profile of Belarus for the coronavirus pandemic.
Source: Belarus Ministry of Health
Short description: The number of tests performed.
Latest estimate: 0.87 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The Belarus Ministry of Health provides daily press releases that report the cumulative number of tests conducted to date. It is unclear whether the reported figures include pending tests.
Prior to 14 April 2020, the Ministry reported imprecise testing figures (e.g. “…more than 64 thousand tests…”). We include these imprecise figures so that our time series extends back to 3 March 2020.
The earliest reported figure that we have been able to find is from 3 March 2020, at which point approximately 5,000 cumulative tests were reported. The number of tests are cumulative since late January 2020, but the press releases do not specify the exact date on which the first test was conducted.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Belgium for the coronavirus pandemic.
Source: Sciensano (Belgian institute for health)
Short description: The number of tests performed.
Latest estimate: 1.64 daily tests per thousand people (as of 22 August 2020).
Detailed description:
Sciensano provides a daily time series of the number of tests conducted per day, which is updated retrospectively as new data becomes available each day.
According to the dataset codebook, dates for new testing figures correspond to the date of laboratory diagnosis (or when not available, date of sampling).
It is unclear whether the testing figures include technologies other than PCR tests. In a communication on 3 April 2020, Sciensano indicates that an antigen test made by Coris Bioconcept may now be used for COVID-19 detection. The communication states that positive antigen test results do not have to be confirmed by a PCR test, but that negative or doubtful cases must be confirmed by a PCR test.
Official figures reported in Sciensano’s daily epidemiological bulletins do not help to clarify whether non-PCR tests are included in the daily time series figures. For example, as of 10 May 2020, the official time series dataset reported a cumulative total of 465,201 tests performed between 1 March and 9 May 2020. However, the epidemiological bulletin for 10 May 2020 states that 325,796 cumulative tests had been performed by laboratories between the beginning of March and 9 May 2020, while an additional 240,305 tests had been performed through the national testing platform (566,101 total tests).
The bulletin indicates that these figures include both PCR and antigen tests, suggesting that the large discrepancy between the 465,201 figure and the combined 566,101 figure may be due to the fact that the official time series data only includes PCR tests, whereas the official bulletin figures include both PCR and antigen tests. However, we have been unable to find official documentation from Sciensano that resolves this ambiguity.
→ View the country profile of Bolivia for the coronavirus pandemic.
Source: Ministry of Health
Short description: The number of tests performed.
Latest estimate: 0.27 daily tests per thousand people (as of 20 August 2020).
Detailed description:
Official testing data from the Bolivian Ministry of Health is reported in daily briefs on their website. The briefs often (but do not always) provide a breakdown of the cumulative number of confirmed, suspected, and negative (also called discarded) tests to date. We measure the total number of tests performed to date as the sum of the number of confirmed and negative tests.
Because of the way the daily briefs report the number of negative tests (“pruebas negativas”) alongside the number of positive cases (“casos positivos”), it may be the case that the number of tests performed is equivalent to the number of people tested.
The reported number of tests performed may include non-PCR tests. Official protocol for COVID-19 laboratory diagnosis (updated in May 2020) indicate that antibody (non-PCR) tests are an important aspect of the country’s screening and diagnostic process (e.g. see pages 24-27 and annex 2). But the source does not explicitly state whether these non-PCR tests are used to confirm or discard suspected cases without a corresponding PCR test result.
As of 5 August 2020, the most recent daily briefs consistently report testing figures in terms of the cumulative number of tests (“pruebas”) performed. However, in the initial months of the COVID-19 outbreak, the daily briefs previously reported testing figures using inconsistent terminology such as samples (“muestras”) tested, people who have undergone tests (”personas, que fueron sometidas a pruebas”), and tests (“pruebas”) performed. It does not appear that this inconsistent use of terminology reflects a substantive change in the reported figures, since there are no corresponding large breaks in the time series.
In 10 briefs the number of negative tests was not reported, preventing us from calculating a total for that date. For these dates with missing official data, we use data provided in this unofficial GitHub repository, which we have cross-referenced against the official data for all dates.
Our data for this series is sourced from a non-official repository of official data. As explained in our FAQ here we regularly audit the accuracy of this repository against direct official channels. Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Brazil for the coronavirus pandemic.
Source: Brazil Ministry of Health
Short description: The number of tests performed.
Latest estimate: 0.23 daily tests per thousand people (as of 15 August 2020).
Detailed description:
The Ministry of Health press releases published on its website intermittently include figures for the number of PCR tests carried out for COVID-19.
The most recent figures explicitly relate to the number of PCR tests carried out in public laboratories. Two earlier press releases – for 20 April 2020 and 7 April 2020 are not explicit as to whether the figures they provide include or excluded private laboratories. The two April releases imply that the figure provided relates to the cumulative since 16 February 2020, stating that ‘Tests for coronavirus began to be carried out from February 16 in public and private laboratories’.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
Brazil includes positive antibody tests in their figures for confirmed cases. Our testing figures—which exclude antibody tests—are not an appropriate comparison in these instances: on this basis there could be more cases than tests, which is not possible. For this reason, we do not calculate the positive rate or number of tests per confirmed case for Brazil.
→ View the country profile of Bulgaria for the coronavirus pandemic.
Source: Bulgaria COVID-19 Information Portal
Short description: The number of tests performed.
Latest estimate: 0.75 daily tests per thousand people (as of 24 August 2020).
Detailed description:
Bulgaria’s official COVID-19 information portal provides irregular updates of the cumulative number of ‘PCR tests done’.
Using web archives we reconstruct the testing time series starting from 11 April 2020. We cannot say with certainty when testing began, only that the earliest observation available to us begins from 11 April 2020. For 19 April 2020, we take the figure provided in Bulgaria’s COVID-19 dashboard as no snapshot was available using web archives. The test figures provided in the dashboard match the figures provided by the information portal for all other dates available.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Canada for the coronavirus pandemic.
Source: Government of Canada
Short description: The number of people tested.
Latest estimate: 1.28 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Government of Canada provides daily updates of the COVID-19 outbreak here, which report the cumulative number of people tested to date.
We construct a daily time series of the cumulative number of people tested to date using Web Archive snapshots of these daily updates. The figures include positive and negative test results, while excluding pending test results.
The figures relate to diagnostic testing completed in laboratories.
We only report data since 18 March 2020 due to a large jump in the time series that occurred between 17 March 2020 (1,018 people tested) and 18 March 2020 (53,975 people tested). We suspect this jump was the result of a backlog in tests waiting to be processed, but the Government of Canada website does not provide an explanation. For comparability over time, we therefore exclude data prior to 18 March 2020.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Chile for the coronavirus pandemic.
Source: Government of Chile Coronavirus information page
Short description: The number of tests performed
Latest estimate: 1.39 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Government of Chile release daily reports that include cumulative and daily totals for the number of PCR tests performed across private and public medical establishments. This data is collected by volunteers and published on GitHub. We take our figures from this GitHub, which we regularly audit for accuracy.
Our data for this series is sourced from a non-official repository of official data. As explained in our FAQ here we regularly audit the accuracy of this repository against direct official channels. Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Colombia for the coronavirus pandemic.
Source: National Institute of Health
Short description: The number of samples tested.
Latest estimate: 0.62 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The Colombian National Institute of Health publishes a dashboard on COVID-19, updated daily. Page 4 of the embedded report includes a graph on “histórico de muestras” (sample history), with a cumulative number of “muestras procesadas” (samples processed).
The report says that “the daily sample chart is based on information loaded by all laboratories that perform SARS-CoV2 diagnostics. At this time, some laboratories are completing the dates for conducting the tests, so the graph will vary as these data are completed.”
Because of this, our time series does not include samples marked as “S.F.” (“Sin Fecha”) in the leftmost bar of the graph; these can represent a high number of samples (3867 as of 21 April 2020).
→ View the country profile of Costa Rica for the coronavirus pandemic.
Source: Costa Rican Ministry of Health
Short description: The number of people tested.
Latest estimate: 0.32 daily tests per thousand people (as of 22 August 2020).
Detailed description:
The Costa Rican Ministry of Health provides daily updates that report confirmed cases, deaths and test results.
These updates state the daily number of people who have been tested – given as the number of confirmed cases (‘Confirma contagio’) and number of people with negative results (‘Descarta contagio’) (thus the figures do not include pending tests). We use these daily updates to construct a full time series of the cumulative number of persons tested to date. These figures have been collected and are made available for download by the news website El Observador, from which we take our data.
Daily testing figures are only available since 11 March 2020. We therefore do not know the first date of testing, or daily figures prior to this date.
For the number of new tests reported on 25 March 2020, we noticed a discrepancy between data published by El Observador and the official data reported by the Ministry of Health. We therefore replaced the number of discarded tests for that day (5) with the official number (65) (official reports: 24 March 2020, 25 March 2020).
Prior to 24 July 2020, suspected COVID-19 cases in Costa Rica could only be confirmed via a positive PCR test result, as described in the 12 June 2020 national guidelines for COVID-19 surveillance. However, revised guidelines published on 24 July 2020 expand the definition of a confirmed COVID-19 case to allow for cases to be confirmed without a PCR test:
“Confirmed case refers to one of the following 2 options:
Laboratory confirmed case: refers to a person who has had the virus that causes the COVID 19 disease regardless of its clinical signs and symptoms. This confirmation will be done by public and private laboratories that have RT-PCR capable of identifying SARS – CoV-2 and have an authorization from an external regulatory entity such as FDA or its equivalent duly authorized by the Ministry of Health as well as the National Center for Frankincense Virology.
By epidemiological link: persons residing at the address of a laboratory confirmed case who develop respiratory symptoms within 21 days after the first day of isolation. (For these people, it will not be necessary to carry out the PCR-RT test)” (translated)
For this reason, we do not know how many confirmed cases since 24 July 2020 have been confirmed via “epidemiological link” (i.e. without a PCR test). Hence, the time series we construct of the cumulative number of persons tested to date may include a number of confirmed cases that never received a PCR test.
This also means that our testing figures are not an appropriate comparison with cases in these instance. For this reason, we do not calculate the positive rate or number of tests per confirmed case for Costa Rica.
Our data for this series is sourced from a non-official repository of official data. As explained in our FAQ here we regularly audit the accuracy of this repository against direct official channels. Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
The Costa Rica testing data includes non-PCR tests, which reduces its comparability with other countries.
→ View the country profile of Cote d’Ivoire for the coronavirus pandemic.
Source: Ministry of Health and Public Hygiene
Short description: The number of samples tested.
Latest estimate: 0.04 daily tests per thousand people (as of 23 August 2020).
Detailed description:
Since 13 April 2020, the Ministry of Health and Public Hygiene publishes the cumulative total of samples taken in a daily post on its official Facebook page. For a small number of daily updates, only the daily change is stated; in these cases, we derive the cumulative total based on the previous day’s cumulative total + daily change.
The updates use the words “samples taken” to describe these figures, which likely indicates that they include tests whose results are still pending.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Croatia for the coronavirus pandemic.
Source: Government of Croatia
Short description: The number of people tested.
Latest estimate: 0.49 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Government of Croatia provides daily updates of confirmed cases, deaths, and testing figures here. We construct a time series based on the cumulative figures reported in web archives of the latest koronavirus.hr press release. It is unclear whether the reported figures include tests for which the results are pending.
Prior to 24 March 2020, testing figures were reported in terms of “samples tested” (e.g. “Ukupno je 317 testiranih uzoraka…”). From 24 March to 6 May 2020, testing figures were reported in terms of “tests performed” (e.g. “Dosad je obavljeno ukupno 3.159 testiranja”). Since 6 May 2020 they have been reported in terms of “people tested” (e.g. “Dosad je testirano ukupno 41.053 osoba”). It does not appear that this change in terminology reflects a substantive change in the figures that are reported, since there are no large breaks in the time series around these changes in terminology.
We have found testing data dating back to 3 March 2020, at which point 247 samples had been tested to date.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Cuba for the coronavirus pandemic.
Source: Ministry of Public Health
Short description: The number of tests performed.
Latest estimate: 0.42 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The Ministry of Public Health publishes data on its official website, which is collected and republished by the COVID-19 Cuba Data project in a dashboard and on GitHub. It includes a time series for the number of tests.
The context in which the figures are published suggests that the number of tests performed may be equivalent to the number of individuals tested. A breakdown into positive and negative results is given, along with a positive test rate. It is also unclear whether the reported figures include tests for which the results are pending.
Our data for this series is sourced from a non-official repository of official data. As explained in our FAQ here we regularly audit the accuracy of this repository against direct official channels.
→ View the country profile of Czech Republic for the coronavirus pandemic.
Source: Ministry of Health
Short description: The number of tests performed.
Latest estimate: 0.62 daily tests per thousand people (as of 22 August 2020).
Detailed description:
The Ministry of Health publishes a data set containing incremental and cumulative daily numbers of COVID-19 tests performed according to laboratory reports. It is updated daily and accessible in CSV and JSON format. No other information about the figures could be found.
→ View the country profile of Democratic Republic of Congo for the coronavirus pandemic.
Source: DRC COVID-19 Pandemic Response Multisectoral Committee
Short description: The number of samples tested.
Latest estimate: 0 daily tests per thousand people (as of 20 August 2020).
Detailed description:
The DRC COVID-19 Pandemic Response Multisectoral Committee publishes daily bulletins on their official Twitter page. These bulletins include the daily new samples tested (“échantillons testés”), but no cumulative total; thus it is unclear how many samples have been tested to date. No other information is provided about the reported testing numbers, such as whether they include pending or non-PCR tests.
→ View the country profile of Denmark for the coronavirus pandemic.
Source: Statens Serum Institut
Short description: The number of tests performed.
Latest estimate: 5.25 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The Danish Statum Serum Institute provides epidemiological update reports on confirmed cases, deaths, testing and demographic details. An archive of .zip files is available on its website. Each contains a .csv file (Test_pos_over_time.csv) with the time series of daily (“Tested”) and cumulative (“Tested_kumulativ”) tests performed, going back to 27 January 2020.
→ View the country profile of Ecuador for the coronavirus pandemic.
Source: Government of Ecuador
Short description: The number of people tested.
Latest estimate: 0.15 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The Government of Ecuador publish daily updates in the form of situation reports and summary infographics. These report the number and status of confirmed cases, deaths and number of samples tested. This data is available daily from 18 March 2020; reports and infographics prior to this date do not include the number of samples tested. But all figures are dated cumulative since 29 February 2020.
The source reports the number of confirmed (‘confirmados’) and negative (‘descartados’) cases, which we sum to get the number of cases tested.
On 24 April 2020, the number of tests suddenly jumped from 23,383 to 45,857, because of what we assume to be the inclusion of rapid tests (“pruebas rápidas”), as made clear by the subsequent infographic published on 27 April 2020. We therefore do not include the 24 April 2020 infographic in our time series; and from 27 April 2020 onwards, we include only PCR tests.
On 11 May 2020, the Government of Ecuador published a report detailing “a reclassification of the records by identity card of the persons and not by the number of tests that have been carried out”. This suggests the government is moving towards reporting figures on the number of people tested, where previously they reported the number of cases tested. The reclassification partially explains the fall in the cumulative total after 4 May 2020.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
Ecuador includes positive antibody tests in their figures for confirmed cases. Our testing figures—which exclude antibody tests—are not an appropriate comparison in these instances: on this basis there could be more cases than tests, which is not possible. For this reason, we do not calculate the positive rate or number of tests per confirmed case for Ecuador.
→ View the country profile of El Salvador for the coronavirus pandemic.
Source: Government of El Salvador
Short description: The number of tests performed.
Latest estimate: 0.38 daily tests per thousand people (as of 22 August 2020).
Detailed description:
The government of El Salvador publishes an online dashboard that reports the number of tests performed (“pruebas COVID19 realizadas hasta hoy”). No information is given on the geographical scope and number of labs included.
The official dashboard only provides a daily snapshot of the number of tests performed today, making it difficult to construct a historical time series. We construct a daily time series dating back to 10 April 2020 of the cumulative number of tests performed using data posted on President Nayib Bukele’s official Facebook page, which match the daily snapshot figures reported in the official dashboard.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Estonia for the coronavirus pandemic.
Source: Estonian Health Board
Short description: The number of tests performed (“Testide koguarv”)
Latest estimate: 0.66 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The Estonian Health Board (“Terviseamet”) is the source of and links to testing data visualized by Open Data Estonia on the Koroonakaart dashboard. The dashboard contains a time series of tests per day and cumulative tests dating back to 25 February 2020, among other data. The Health Board provides further description of the data, including noting that past data may be revised if test results are retroactively corrected or new results are received.
→ View the country profile of Ethiopia for the coronavirus pandemic.
Source: Ethiopian Public Health Institute
Short description: The number of tests performed.
Latest estimate: 0.18 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The Ethiopian Public Health Institute in collaboration with the Ethiopian Ministry of Health provides daily press releases of the number of tests performed to date. It is unclear whether the reported figures include tests for which the results are pending.
The earliest record we have found is for 17 March 2020, at which point 79 tests had been performed to date. It is unclear when the first test was performed.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Fiji for the coronavirus pandemic.
Source: Fiji Ministry of Health & Medical Services
Short description: The number of tests performed.
Latest estimate: 0.04 daily tests per thousand people (as of 16 August 2020).
Detailed description:
The Fiji Ministry of Health & Medical Services provides a series for the number of tests per day, the number of tests per week, and the cumulative number of tests conducted to date.
The source notes “Laboratory testing for COVID-19 in Fiji began on 28 January 2020 with samples shipped to the WHO collaborating center reference laboratory – the Victorian Infectious Disease Reference Laboratory (VIDRL) at the Peter Doherty Institute for Infection and Immunity in Melbourne, Australia. Local testing for COVID-19 started at the Fiji Centre for Disease Control (Fiji CDC) on 11 March 2020 using the gold standard method for COVID-19 testing – realtime RT-PCR. Samples are sent from health facilities around the country to Fiji CDC for testing.”
→ View the country profile of Finland for the coronavirus pandemic.
Source: Finnish Department of Health and Welfare COVID-19 data dashboard
Short description: The number of samples tested.
Latest estimate: 1.35 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The Finnish Department of Health and Welfare publishes a dashboard of confirmed cases, deaths and samples tested. The dashboard provides both daily and cumulative test numbers. We extract the complete daily time series using the official API.
Further descriptions of the testing data are provided here. In this document it notes that:
– “Public and private sector laboratories report to THL (health authority) the number of samples tested per hospital district”
– “The actual number of infections in Finland is probably higher than reported, as not all mild symptoms have been tested so far and no information is available on the number of asymptomatic infections.”
– “Coronavirus testing is performed in both the public and private sectors in several laboratories.”
The data description notes that tests are being rationed and allocated according to the following strategy: “Samples are taken primarily from patients with severe symptoms of respiratory infection and from health and social care staff to ensure staff health and labor adequacy….Attention has also been paid to vulnerable groups of patients with some underlying disease and to those over 70 years of age.”
This data series extends to the 30th January, when test figures were very low (73 tests per day). It is likely this is was the first day, or close to the first day of testing.
→ View the country profile of France for the coronavirus pandemic.
Source #1: National Public Health Agency
Short description: The number of people tested.
Latest estimate: 1.43 daily tests per thousand people (as of 21 August 2020).
Detailed description:
Since 13 May 2020, the National Public Health Agency has replaced its previous weekly updates of the number of tests performed with a daily-updated file with the count of people tested, available on the French government’s open data portal.
The data covers all results of test laboratories (RT-PCR) carried out by all city laboratories and hospitals concerning SARS-COV2. The source notes that “the time to report back tests can exceed 9 days in some cases. The indicators are adjusted daily according to the reception of the results.”
This data does not allow us to publish any cumulative total, since no information is available on how many people had been tested prior to the first day of reporting.
Source #2: National Public Health Agency
Short description: The number of tests performed.
Latest estimate: 0.23 daily tests per thousand people (as of 5 May 2020).
Detailed description:
The National Public Health Agency published updates for the number of tests performed until early May 2020. The figures in the time series relate to tests performed since 24 February 2020. Since the 24 March 2020 update, in addition to the previously reported hospital tests, laboratory tests are also reported. Only the total number of laboratory tests performed between 24 February and 24 March 2020 is presented, without a time series. Our figure for 24 March 2020 adds the cumulative number of laboratory tests since 24 February 2020 (6,500) to the hospital tests figure (101,046). As such 24 March 2020 represents a break in the series.
The report dated 21 May 2020 specified a new screening information system (SI-DEP) had been deployed from week 20 (11-17 May 2020) and that epidemiological trends and test positivity rates would be available from 28 May 2020. The report published in late May reported the number of people tested instead of the number of tests performed, we therefore created a second time series for France.
→ View the country profile of Germany for the coronavirus pandemic.
Source: Robert Koch Institut
Short description: The number of tests performed. Note that the number of laboratories reporting in the data seems to vary from week to week.
Latest estimate: 1.49 daily tests per thousand people (as of 16 August 2020).
Detailed description:
To determine how many laboratory tests regarding SARS-CoV-2 are carried out per calendar week in Germany and how many tests are positive or negative, the RKI has started a Germany-wide laboratory query. However, the number of laboratories reporting data seems to vary from week to week.
The report published on 19 August 2020 states that “from the beginning of the collection up to and including calendar week 33/2020”:
– The cumulative total of samples tested was 10,197,366;
– For calendar week 33 (which ends 16 August), 181 labs reported 875,524 samples tested;
– For calendar week 32 (which ends 9 August), 165 labs reported 730,300 samples tested;
– For calendar week 31 (which ends 2 August), 165 labs reported 578,099 samples tested;
– For calendar week 30 (which ends 26 July), 180 labs reported 570,746 samples tested;
– For calendar week 29 (which ends 19 July), 176 labs reported 538,229 samples tested;
– For calendar week 28 (which ends 12 July), 178 labs reported 510,103 samples tested;
– For calendar week 27 (which ends 5 July), 149 labs reported 504,082 samples tested;
– For calendar week 26 (which ends 28 June), 179 labs reported 466,459 samples tested;
– For calendar week 25 (which ends 21 June), 175 labs reported 387,484 samples tested;
– For calendar week 24 (which ends 14 June), 172 labs reported 326,645 samples tested;
– For calendar week 23 (which ends 7 June), 176 labs reported 340,986 samples tested;
– For calendar week 22 (which ends 31 May), 178 labs reported 405,269 samples tested;
– For calendar week 21( which ends 24 May), 179 labs reported 353,467 samples tested;
– For calendar week 20 (which ends 17 May), 183 labs reported 432,666 samples tested;
– For calendar week 19 (which ends 10 May), 182 labs reported 403,875 samples tested;
– For calendar week 18 (which ends on 3 May), 175 labs reported 326,788 samples tested;
– For calendar week 17 (which ends on 26 April), 178 labs reported 363,890 samples tested;
– For calendar week 16 (which ends on 19 April), 168 labs reported 331,902 samples tested;
– For calendar week 15 (which ends on 12 April), 164 labs reported 380,197 samples tested;
– For calendar week 14 (which ends on 5 April), 154 labs reported 408,348 samples tested;
– For calendar week 13 (which ends on 29 March), 151 labs reported 361,515 samples tested;
– For calendar week 12 (which ends on 22 March), 152 labs reported 348,619 samples tested;
– For calendar week 11 (which ends on 15 March), 114 labs reported 127,457 samples tested.
– Up to and including calendar week 10 (which ends on 8 March), 90 labs reported 124,716 samples tested.
By subtracting each weekly change from the cumulative total, we retrospectively work out the cumulative totals by the end of each week.
Since laboratories can post-check the tests of past calendar weeks in the RKI test number query, previous figures may be revised upwards slightly in subsequent reports. The source is explicit that these figures refer to tests performed and that this will not equal the number of people tested, because of multiple tests per person.
→ View the country profile of Ghana for the coronavirus pandemic.
Source: Outbreak Response Management
Short description: The number of samples tested.
Latest estimate: 0.04 daily tests per thousand people (as of 21 August 2020).
Detailed description:
Ghana’s Outbreak Response Management provides daily situation updates on the total number of samples tested, often provided with a breakdown of the number of samples tested through routine surveillance, contact tracing, and mandatory quarantine.
Between 10 and 26 May 2020, the daily situation updates provide figures for both the number of samples tested and the number of persons tested. Over this period the two figures differ from one another by exactly 981 tests on each day, which is due to the fact that many of the 1,041 cumulative persons tested under mandatory quarantine were tested multiple times (2,022 samples tested on 1,041 persons). This discrepancy of 981 tests is small, representing less than 1% of cumulative samples tested at the time. In contrast, the cumulative number of samples tested through routine surveillance and contact tracing are exactly equal to the cumulative number of persons tested on each day in these respective categories.
Prior to 10 May 2020, the daily situation updates provide testing figures using inconsistent terminology, varying between “samples tested”, “persons tested”, “suspected cases tested”, and similar language. Nevertheless, we choose to include all of these figures within our constructed time series, since the variation in terminology does not appear to correspond to large breaks in the time series. In addition, based on the fact that the number of samples tested differs only slightly from the number of persons tested over the 10 May to 26 May 2020 period, we have no reason to believe that the variation in terminology prior to 10 May has any serious ramifications for the interpretation of the time series as the cumulative number of samples tested.
From 10 May 2020 onwards, the daily situation updates consistently report the cumulative number of samples tested.
→ View the country profile of Greece for the coronavirus pandemic.
Source: The National Organization of Public Health
Short description: The number of samples tested.
Latest estimate: 1.8 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Greek National Organization of Public Health publishes daily COVID-19 press releases detailing the number of confirmed cases, deaths and samples tested.
The report refers to the number of clinical samples that have been tested to date since 1 January 2020. This includes the number of samples that tested either positive or negative. Reports are also explicit that the figures relate to samples tested and not the number of people, with figures “including more than one sample per person tested” (“συμπεριλαμβάνονται και περισσότερα από ένα δείγματα ανά άτομο που ελέγχθηκε”).
It is not clear whether the data covers all testing in Greece as the source report notes these are “Samples that have been tested in the laboratories cooperating with EODY” (“Δείγματα που έχουν ελεγχθεί στα συνεργαζόμενα με τον ΕΟΔΥ εργαστήρια”).
The archived official website provides a list of daily reports. The earliest archived report available is from 20 March 2020 when 7,172 clinical samples had been tested since 1 January 2020 with the most recent dated 30 May 2020. From 3 April 2020, daily reports are available via the Greek National Organization of Public Health’s COVID-19 Announcements and press releases.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Hong Kong for the coronavirus pandemic.
Source: Centre for Health Protection, Department of Health
Short description: The number of tests performed.
Latest estimate: 1.64 daily tests per thousand people (as of 11 August 2020).
Detailed description:
The Hong Kong Centre for Health Protection, Department of Health provides monthly numbers of the COVID-19 viral tests performed. We calculate the cumulative number of tests performed each month in our series. The last update with provisional data is up until 14 April 2020, this information is updated once a week by the Centre for Health Protection.
Tests performed are conducted by the Public Health Laboratory Services Branch of the Department of Health and Hospital Authority.
The cumulative total begins from 1 January 2020.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Hungary for the coronavirus pandemic.
Source: Government of Hungary
Short description: The number of samples tested in an accredited laboratory.
Latest estimate: 0.26 daily tests per thousand people (as of 24 August 2020).
Detailed description:
A figure for the ‘number of samples tested in an accredited laboratory’ (“akkreditált laboratóriumban vizsgált minták száma”) is provided in a graphic in the govenrment’s coronavirus information website. No other information about the data is provided.
We use these official data as collected by the visual and data journalism team of Budapest-based investigative center Atlatszo, made accessible in a public spreadsheet. We have cross-checked a sample of the figures in this unofficial spreadsheet against official figures.
Our data for this series is sourced from a non-official repository of official data. As explained in our FAQ here we regularly audit the accuracy of this repository against direct official channels. Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Iceland for the coronavirus pandemic.
Source: Government of Iceland
Short description: The number of tests performed.
Latest estimate: 1.3 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The Icelandic Office of Public Health publishes a complete time series of tests performed, broken down into two labs where these samples are tested. No other information is provided. It is not clear whether these figures include samples that are pending test results. The daily time-series data only extends back to 27 February 2020.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of India for the coronavirus pandemic.
Source #1: Indian Council of Medical Research
Short description: The number of people tested.
Latest estimate: 0.02 daily tests per thousand people (as of 23 April 2020).
Detailed description:
The ICMR reports separate figures for both “samples tested” and “people tested” at press conferences and in press releases (shown separately in the charts above). No other details are provided.
The press releases from ICMR do not always stay online for very long. The reason for this is unknown, but the releases are being backed up at this GitHub repository.
On some occasions there appear to have been more than one update released per day. Where we are aware of multiple observations for the day, we show the number for the earlier release.
The ICMR website does not explicitly state whether the reported figures refer to PCR tests only. From contextual information, it appears that the reported figures may also include samples that were tested using a TrueNat non-PCR test. ICMR communications on 21 May 2020 and 19 April 2020 indicate that TrueNat tests are being used in diagnostic testing. These TrueNat tests likely account for a small minority of all samples tested.
Our data for this series is sourced from a non-official repository of official data. As explained in our FAQ here we regularly audit the accuracy of this repository against direct official channels. Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
Source #2: Indian Council of Medical Research
Short description: The number of samples tested.
Latest estimate: 0.62 daily tests per thousand people (as of 22 August 2020).
Detailed description:
The ICMR reports separate figures for both “samples tested” and “people tested” at press conferences and in press releases (shown separately in the charts above). No other details are provided.
The press releases from ICMR do not always stay online for very long. The reason for this is unknown, but the releases are being backed up at this GitHub repository.
On some occasions there appear to have been more than one update released per day. Where we are aware of multiple observations for the day, we show the number for the earlier release.
The ICMR website does not explicitly state whether the reported figures refer to PCR tests only. From contextual information, it appears that the reported figures may also include samples that were tested using a TrueNat non-PCR test. ICMR communications on 21 May 2020 and 19 April 2020 indicate that TrueNat tests are being used in diagnostic testing. These TrueNat tests likely account for a small minority of all samples tested.
Our data for this series is sourced from a non-official repository of official data. As explained in our FAQ here we regularly audit the accuracy of this repository against direct official channels. Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Indonesia for the coronavirus pandemic.
Source: Emerging infections, Indonesian Ministry of Health
Short description: The number of people tested.
Latest estimate: 0.49 daily tests per thousand people (as of 20 August 2020).
Detailed description:
The Indonesian Ministry of Health updates its COVID-19 dashboard twice a day but doesn’t keep time series of past data. We found past values using Internet Archive’s Wayback Machine. The two dashboard URLs (1, 2) seem to lead to the exact same dashboard.
The dashboard shows the latest number of “Kasus dg Spesimen Diperiksa”, which translates to “Cases with Specimens Checked”.
Official diagnostic guidelines published on 23 March 2020 provide some reason to be concerned that serological (antibody) tests are included in the cumulative testing figures, stating that:
“Handling of COVID-19 in Indonesia uses antibodies and / or antigen Rapid Tests (RT) in case of contact from a positive patient. RT antibodies are also used for case detection…in areas that do not have facilities for RT-PCR inspection. RT antibody examination results are confirmed using RT-PCR.” (translated)
However, testing figures provided by the National Agency for Disaster Countermeasure (BNPB) via twitter explicitly report the number of PCR tests conducted to date, which match the cumulative testing figures provided in the Ministry of Health dashboard. The PCR testing figures reported by the BNPB include a small number of “rapid molecular tests” (“Test Cepat Molekular”), which are diagnostic tests not to be confused with serological (antibody) tests. Of 998,406 cumulative people tested as of 11 August 2020, 26,184 (2.6%) were tested via rapid molecular testing (source). These rapid molecular testing figures are included in the time series we construct.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Iran for the coronavirus pandemic.
Source: Government of Iran
Short description: The number of tests performed.
Latest estimate: 0.3 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Government of Iran provides daily press releases of the cumulative number of tests performed to date. It is unclear whether the reported figures include tests for which the results are pending.
We have found testing data dating back to 5 April 2020, at which point 189,790 tests had been conducted to date. It is not clear when the first test was conducted.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Iraq for the coronavirus pandemic.
Source: Iraq Ministry of Health and Environment
Short description: The number of samples tested.
Latest estimate: 0.5 daily tests per thousand people (as of 19 August 2020).
Detailed description:
The Iraq Ministry of Health and Environment provides daily COVID-19 reports on their official website that include the daily and cumulative number of samples tested. These reports date back to late March 2020 but did not begin including testing numbers until 7 April 2020; the cumulative total then was already 26,331. The testing numbers are described as being from “all the specialized laboratories in Iraq” (translated from Arabic). Other reports on the website suggest that these testing numbers are for PCR tests only, but we could not find definitive confirmation of this. No other information is provided.
→ View the country profile of Ireland for the coronavirus pandemic.
Source: Government of Ireland
Short description: The number of tests performed.
Latest estimate: 1.51 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Government of Ireland has released a data hub on 18 June 2020 that provides updates on the number of tests completed. The earliest observation is from 18 March 2020, however, the exact date these cumulative figures date back to is unclear.
The total we report is the sum of total tests completed in hospitals and all other labs (NVRL and Cherry Orchard).
The source notes “Data presented are reported by the Health Protection Surveillance Centre (HPSC) at 3pm each day (data as of midnight previous day). This data will be updated on a daily basis from Monday to Saturday. As some cases are tested multiple times over the duration of the illness (with positive results) the total number of positive tests does not correspond to the confirmed COVID-19 cases reported by the HPSC.”
→ View the country profile of Israel for the coronavirus pandemic.
Source: Ministry of Health
Short description: The number of tests performed.
Latest estimate: 2.5 daily tests per thousand people (as of 17 August 2020).
Detailed description:
The Israel Ministry of Health publishes a coronavirus dashboard on its website. It includes a downloadable data file which includes the cumulative number of coronavirus detection tests conducted. This data file is updated daily and includes data up to one week before publication.
The first observation in the time series is from 26 January 2020 which covers most, if not all, of the relevant period.
No further information on the geographical scope, number of labs, or types of test included are known.
The MOH also publishes COVID-19 updates on its official channel on Telegram. However, this data is published in a format that is extremely challenging to collect. We previously relied on the data collected and made available on GitHub. On 19 April 2020, the person who maintains the GitHub repository confirmed to us that the units refer to the number of tests performed, after checking the information with the Ministry of Health. However, on 29 May 2020 we became aware of the easily accessible Israel MOH series which we now report.
→ View the country profile of Italy for the coronavirus pandemic.
Source #1: Ministero della Salute
Short description: The number of people tested.
Latest estimate: 0.7 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The official data provided by the Ministry of Health is compiled by the Department of Civil Protection on GitHub. Figures relate to the total number of people tested. This figure is available for each individual region separately at the source indicated. However it appears that data is missing for individual regions for particular dates – we assume that these are also missing from the aggregate total for Italy that we provide. The list of missing data, in English, can be seen at the bottom of this data dashboard built by Franco Mossotto.
It is clear that there are delays in timeline running up to a test being reported – both in terms of the time it takes for a symptomatic person to receive a test, and in the time for that test to then get reported in the data. For one region, Lombardy, an investigation found the latter delay in the early period of the outbreak to be around 3-4 days.
There is significant variation in testing practices – in terms of how many tests are conducted and how they are allocated across the population – across regions and these practices have also changed over time. The extent to which tests pending results are included appears to vary across regions.
Pietro Monticone and Riccardo Valperga have written a very helpful and detailed description of these data quality issues here in GitHub.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
Source #2: Ministero della Salute
Short description: The number of tests performed.
Latest estimate: 1.1 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The official data provided by the Ministry of Health is compiled by the Department of Civil Protection on GitHub. Figures relate to the total number of tests performed. This figure is available for each individual region separately at the source indicated. However it appears that data is missing for individual regions for particular dates – we assume that these are also missing from the aggregate total for Italy that we provide. The list of missing data, in English, can be seen at the bottom of this data dashboard built by Franco Mossotto.
It is clear that there are delays in timeline running up to a test being reported – both in terms of the time it takes for a symptomatic person to receive a test, and in the time for that test to then get reported in the data. For one region, Lombardy, an investigation found the latter delay in the early period of the outbreak to be around 3-4 days.
There is significant variation in testing practices – in terms of how many tests are conducted and how they are allocated across the population – across regions and these practices have also changed over time. The extent to which tests pending results are included appears to vary across regions.
Pietro Monticone and Riccardo Valperga have written a very helpful and detailed description of these data quality issues here in GitHub.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Japan for the coronavirus pandemic.
Source #1: Ministry of Health, Labor and Welfare
Short description: The number of people tested.
Latest estimate: 0.17 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Ministry of Health, Labor and Welfare issues daily reports for the total number of people tested. Daily changes in the number of people tested are included in parentheses. For 3 April and 21 April 2020, the daily changes do not reconcile with the cumulative totals. Moreover, it is clear that revisions to past data are simply added to the new cumulative total, making the time series of ‘daily tests’ very erratic.
This includes three cases where the cumulative number of people tested falls: (1) 19 March 2020; (2) 25 March 2020; and (3) 15 May 2020. For case (1) see footnote 4 which indicates a past mistake has been noticed and the cumulative figure revised on the date to adjust for this. For case (2), we could not find the associated footnote. For case (3), see footnote 3 of the press release which indicates a past mistake has been noticed and the cumulative figure revised on the date to adjust for this.
It isn’t clear what exact date these cumulative tests date back to, but it is earlier than 10 February 2020 when the source reports 938 people had been tested. Prior to 10 February 2020, the press releases provide reports of coronavirus infections for the Diamond Princess cruise ship quarantined in Yokohama port. We do not report these numbers in the time series as it is unclear how they relate to the cumulative totals.
The MOH’s press release dated 10 July 2020 notes “The Tokyo Metropolitan Government has retroactively changed the number of people who perform tests after July 9, and the number of people after May 13 includes the number of antigen tests in addition to PCR tests. (4,000 or more) is increasing” (translated). In the MOH’s 15 July 2020 press release, footnote 3 mentions that “For some local governments, the number of antigen testers is included.” (translated).
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
The Japan testing data includes non-PCR tests, which reduces its comparability with other countries.
Source #2: Ministry of Health, Labor and Welfare
Short description: The number of tests performed.
Latest estimate: 0.13 daily tests per thousand people (as of 22 August 2020).
Detailed description:
On 11 April 2020, the MOH started providing a daily time series on the “Implementation status of PCR tests for new coronavirus in Japan (based on the date on which results were determined” (via Google translate).
With each update, some daily totals are revised from earlier releases of the time series.
From 20 May 2020, the MOH changed how it reports the number of tests performed. From 18 February to 25 March 2020 (as of 20 May 2020), the MOH reports the number of tests implemented each week, and the daily totals thereafter. We calculate the cumulative total from the weekly and daily figures. There are instances where the cumulative totals reported in the press releases do not sum to the cumulative total calculated from the weekly and daily figures. This includes the series released on 20/21 May and 24/25 May 2020. The source does not provide additional details about this discrepancy.
→ View the country profile of Kazakhstan for the coronavirus pandemic.
Source: Kazakhstan National Center for Public Health
Short description: The number of tests performed.
Latest estimate: 0.82 daily tests per thousand people (as of 14 August 2020).
Detailed description:
The Kazakhstan National Center for Public Health provides an official time series of the total number of tests performed to date. It is unclear whether the reported figures include tests for which the results are pending.
The website explicitly states that “one person could be tested several times”, indicating that the number of tests performed (“Проведено тестов”) is not equivalent to the number of people tested. The reported number of tests performed may include non-PCR tests. Official COVID-19 prevention guidelines indicate that antibody (non-PCR) tests are an important aspect of the country’s screening and diagnostic process (see pages 48-54). But the source does not explicitly state whether these non-PCR tests are included in the reported cumulative number of tests performed to date.
The earliest reported figure is from 13 March 2020, at which point 126 tests had been conducted.
→ View the country profile of Kenya for the coronavirus pandemic.
Source: Ministry of Health
Short description: The number of samples tested.
Latest estimate: 0.08 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The Kenya Ministry of Health provides daily press releases and tweets (@MOH_Kenya), sometimes reporting the number of samples tested in the past 24 hours and sometimes reporting the cumulative number of samples tested to date.
The tweets and press releases usually report testing figures in terms of the number of “samples tested”. However, the wording used in some tweets and press releases makes it unclear whether the figures relate to the number of samples tested or the number of people tested. For example, the relevant tweet on 4 June 2020 says “2,640 samples tested in the last 24 hours”, while the relevant tweet on 1 August 2020 states that “727 people are the latest to test positive for Covid-19 from a sample of 6,371 tested in the last 24 hrs”. It does not appear that this inconsistency in terminology reflects a substantive change in the figures that are reported, since there are no large breaks in the time series that correspond to the use of “samples tested” versus “people tested” terminology. Instead, the wording of the tweets and press releases suggests that the reported number of samples tested may be equivalent to the number of people tested.
It is unclear whether the reported figures include tests for which the results are pending.
We have found testing data dating back to 6 March 2020, at which point 31 tests had been conducted to date.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Kuwait for the coronavirus pandemic.
Source: Kuwait Ministry of Health
Short description: The number of tests performed.
Latest estimate: 0.96 daily tests per thousand people (as of 21 August 2020).
Detailed description:
The Kuwait Ministry of Health provides daily reports of the daily (“NEW TESTS”) and cumulative (“Total”) number of tests performed on their official Twitter account. COVID-19 reports date back to early March 2020 but did not begin including testing numbers until 13 May 2020; the cumulative total then was already 227,000. From 13–29 May 2020, the daily number was termed “NP swab last 24 h” and the cumulative number “Total Investigations.”
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Latvia for the coronavirus pandemic.
Source: Center for Disease Prevention and Control
Short description: The number of tests performed.
Latest estimate: 0.9 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Center for Disease Prevention and Control publishes a timeseries of the number of laboratory tests performed to detect SARS-CoV-2 virus (“Veikto laboratorisko testu skaits, lai noteiktu SARS-CoV-2 vīrusu”) on the Latvian Government’s Open Data Portal. No information is given on the geographical scope and number of labs included.
→ View the country profile of Lithuania for the coronavirus pandemic.
Source: Ministry of Health
Short description: The number of samples tested.
Latest estimate: 1.49 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Ministry of Health provides fairly regular updates of the number of samples that have been analyzed/tested for suspected coronavirus to date (“Iki šiol iš viso ištirta ėminių dėl įtariamo koronaviruso”). It is not clear the exact date these cumulative figures date back to. Using web archives, we construct a time series by looking at earlier snapshots of the website.
The figures exclude samples that remain untested (i.e. tests pending results).
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Luxembourg for the coronavirus pandemic.
Source: Luxembourg Government situation update
Short description: The number of people tested.
Latest estimate: 2.81 daily tests per thousand people (as of 22 August 2020).
Detailed description:
The Luxembourg government provides a complete time series of the cumulative number of people tested since 25 February 2020.
→ View the country profile of Malaysia for the coronavirus pandemic.
Source: Ministry of Health
Short description: The number of people tested.
Latest estimate: 0.27 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Malaysian Ministry of Health provides daily situation updates, including the cumulative number of individuals sampled, available from 19 May 2020. This figure can be found in the infographic titled ‘Jumlah Keseluruhan Taburan Kes COVID-19 di Malaysia’ labelled as ‘Jumlah Individu Disampel’. Since 19 April 2020, we are not aware of this source reporting positive, negative, and pending test figures separately, as was the case before.
The situation updates for 27/28 May 2020 and 2 June 2020 onwards do not include the infographic detailing the total number of people tested. In these instances, we rely on the MOH daily press conference livestreams uploaded to the official facebook page. The sum of positive and negative tests reported in the livestream match the total people tested figures reported in the situation updates.
For 14/15 May 2020, we use the Director General of Health, daily press statement figures. It details the jump in people tested from 14th to 15th May was due to data consolidation efforts and an improved data collection system.
Prior to 19 May 2020, we relied on MOH web archives to reconstruct the time series. The MOH provided daily updates to the total number of cases tested. It is not clear the exact date these cumulative figures date back to. The earliest release we have been able to find begins from the 14 February 2020. This MOH web archive provides a breakdown of the number of positive, negative, and still pending tests. We report total tests as the sum of positive and negative tests, excluding pending tests. The title of the table indicates that these figures relate to cases. As far as we are aware, cases are equivalent to individuals tested.
In a previous version of this page, we reported total tests as the sum of positive, negative, and pending tests. However, since 7 April 2020, the source has not reported the number of pending tests each day. For this reason, we have updated the time series so that total tests is equal to the sum of positive and negative tests (excluding pending tests) for all days on which we report data.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Maldives for the coronavirus pandemic.
Source: Maldives Ministry of Health Official Twitter page
Short description: The number of samples tested.
Latest estimate: 2.03 daily tests per thousand people (as of 21 August 2020).
Detailed description:
The Maldives Health Protection Agency, part of the Ministry of Health, provides daily updates on their official Twitter page (@HPA_MV) with the cumulative total of samples tested. They note that the total includes “repeated samples,” and likely also includes pending tests. This is our source starting 26 June 2020.
Before the 26 June 2020, our source was daily update videos posted to the Maldives Ministry of Health official Facebook page. Toward the end of these videos they report the number of positive, pending, and total “Laboratory Sample[s]” tested; before 16 June 2020 the number of negative samples was also reported. The positive, negative, and total numbers are cumulative, while the pending numbers are current as of that day.
From 16–25 June 2020, we used the reported total, which likely includes pending tests. Before the 16th when negative numbers were reported, we used the total of positive and negative numbers rather than the reported total, since the latter generally included pending values and occasionally contained discrepant numbers that did not match any combination of the positive, negative, or pending numbers. It is not clear when testing first began; data is only available from 16 March 2020 where it was reported that 221 tests had been conducted.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Malta for the coronavirus pandemic.
Source: COVID-19 Malta Public Health Response Team
Short description: The number of tests performed.
Latest estimate: 5.37 daily tests per thousand people (as of 22 August 2020).
Detailed description:
The COVID-19 Malta Public Health Response Team provides updates on the number of tests performed in the last 24 hours and the cumulative total number of tests performed, on its dedicated dashboard and on GitHub. This time series only includes PCR tests. The earliest observation available is from 6 February 2020.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Mexico for the coronavirus pandemic.
Source: Government of Mexico
Short description: The number of people tested.
Latest estimate: 0.07 daily tests per thousand people (as of 19 August 2020).
Detailed description:
The Government of Mexico makes several datasets available along with its COVID-19 dashboard. The files can be downloaded in CSV format.
To arrive at the time series shown on our website, we use the CSV files for confirmed cases (“confirmados”) and negative cases (“negativos”). For each file, we only keep the row that reports national data, merge the two files together, and add up confirmed and negative cases to find the number of people tested each day.
Data starts on 1 January 2020; we do not know if this is because tests started on that date or because earlier data is not available.
→ View the country profile of Morocco for the coronavirus pandemic.
Source: Morocco Ministry of Health
Short description: The number of people tested.
Latest estimate: 0.6 daily tests per thousand people (as of 21 August 2020).
Detailed description:
The Morocco Ministry of Health provides daily updates of the cumulative number of both confirmed cases (“Cas confirmés”) and cases excluded following a negative laboratory result (“Cas exclus suite à un résultat négatif du laboratoire”); we add these two numbers together to derive a cumulative total. We construct a time series of the cumulative total number of cases tested to date using updates from the Ministry of Health’s official Twitter page, (@Ministere_Sante). There are usually two updates per day, and we use the later one. The earliest reported numbers are from 7 February 2020, at which point 9 cases had been tested.
From 2 March to 18 May 2020 we used data stored in this unofficial GitHub repository instead of the official source to automate data collection. We have cross-checked a sample of the figures reported in the unofficial source against official data reported by the Ministry of Health to ensure accuracy.
Our data for this series is sourced from a non-official repository of official data. As explained in our FAQ here we regularly audit the accuracy of this repository against direct official channels. Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Myanmar for the coronavirus pandemic.
Source: Myanmar Ministry of Health and Sports
Short description: The number of samples tested.
Latest estimate: 0.03 daily tests per thousand people (as of 22 August 2020).
Detailed description:
The Myanmar Ministry of Health and Sports provides daily situation reports of the total number of samples tested to date. These figures include positive and negative test results, but exclude pending results.
The earliest reported figure we have been able to find on the ministry website is from 3 April 2020, at which point 1,183 samples had been tested.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Nepal for the coronavirus pandemic.
Source: Ministry of Health and Population
Short description: The number of tests performed.
Latest estimate: 0.4 daily tests per thousand people (as of 19 August 2020).
Detailed description:
The Nepal Ministry of Health and Population publishes daily reports on COVID-19, including the number of RT-PCR tests performed nationally and in different labs.
Before 11 May 2020, the reports differentiated between positive, negative, and pending tests, allowing removal of pending tests from the total. Since 11 May 2020 the number of pending tests is no longer reported and thus it is unclear if they are included in the reported total.
Our data for this series is sourced from a non-official repository of official data. As explained in our FAQ here we regularly audit the accuracy of this repository against direct official channels. Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Netherlands for the coronavirus pandemic.
Source: Dutch National Institute for Public Health and the Environment
Short description: The number of people tested.
Latest estimate: 0.93 daily tests per thousand people (as of 16 August 2020).
Detailed description:
The Dutch National Institute for Public Health and the Environment releases daily epidemiological reports of the COVID-19 situation. The reports contain a weekly time series of the number of persons tested per week since 9 March 2020. The reports also include the number of persons tested per day in the current week.
There appear to be reporting lags, such that the number of labs reporting for days in the current week tend to be lower than for the previous week. As a result, the reported figures are updated retrospectively as additional labs report their results. The figures we display are based on the latest available epidemiological report.
The epidemiological reports state that ‘all laboratories in the Netherlands that perform diagnostics for SARS-CoV-2 have been asked from 9 March 2020 to report these data daily.’ However, the number of labs reporting in a given weekly figure has varied over the course of the reports, which may affect the consistency of the time series we present.
→ View the country profile of New Zealand for the coronavirus pandemic.
Source: New Zealand Ministry of Health
Short description: The number of tests performed.
Latest estimate: 2.94 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The New Zealand Ministry of Health provide a time series for tests per day and cumulative tests. The daily figures begin on 9 March 2020, and the cumulative figures date back to 22 January 2020 when testing began.
A note to the data states that “The total number of tests conducted is greater than the number of people tested, because some people are tested more than once.”
→ View the country profile of Nigeria for the coronavirus pandemic.
Source: Nigeria Centre for Disease Control
Short description: The number of samples tested.
Latest estimate: 0.02 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The Nigeria Centre for Disease Control, on the homepage of the website dedicated to COVID-19, publishes a number of samples tested.
This table often displays figures in a format such as “> 7153”. When this is the case, we have taken the raw number (in our example 7,153) as the cumulative total for that day.
No information is given on the geographical scope and number of labs included. It is also unclear whether the reported figures include tests for which the results are pending.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Norway for the coronavirus pandemic.
Source: Norwegian Institute of Public Health
Short description: The number of people tested
Latest estimate: 1.81 daily tests per thousand people (as of 19 August 2020).
Detailed description:
The Norwegian Institute of Public Health produce daily update reports of confirmed cases, deaths and test results. This information includes demographic information of confirmed cases.
These daily reports state the cumulative number of people who have been tested – including the number and share who tested positive. We can use these daily updates to construct a full time-series. Note that the source states that “Data for the last few days is incomplete and will be updated in upcoming daily reports”, that the “Number of tested and proportion of positive findings among the tested is affected by changes in test criteria”. It also notes that “Laboratory data is now retrieved from the new national laboratory database, which gives us [a] more complete data set with regard to sampling date”. Earlier reports noted that “The figure contains only numbers from laboratories that have reports that include the sampling date” [all via Google Translate]. As such, this seems to imply an increase in coverage across laboratories over time.
In addition to the aggregate number of people tested, the source provides a regional breakdown, in which it is noted that tests in ‘private labs that analyze tests across the country’ are excluded. The sum of this breakdown is lower than the aggregate figure provided. We take this as an indication that the main aggregate figure we report includes all private testing.
Daily reports with figures on testing are only available dating back to 16 March 2020. We therefore do not know the first date of testing, only that as of 16 March 2020, 18,062 people had been tested.
The time of day to which the testing figures relate appears to have changed. Earlier daily reports refers to those tested up until 3pm (local time) of the previous day. More recent reports relate to those tested up until midnight (local time) of the previous day.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Oman for the coronavirus pandemic.
Source: Oman Ministry of Health
Short description: It is unclear whether the number of tests refers to samples or individuals tested.
Detailed description:
The Oman Ministry of Health provides daily statements, including the total number of tests within the last 24 hours. It is unclear whether tests refer to samples or individuals tested.
The earliest observation on testing figures begins from 4 June 2020. No cumulative total is provided by the source, therefore this time series is only visible on our graphs for daily tests.
The daily statements issued by the MOH do not provide test figures after 6 August 2020.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Pakistan for the coronavirus pandemic.
Source: Government of Pakistan
Short description: The number of tests performed.
Latest estimate: 0.11 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The government of Pakistan publishes a dashboard showing up-to-date national data.
The figures sometimes shows important 1-day increases due to the inclusion of more labs.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Panama for the coronavirus pandemic.
Source: Panama Ministry of Health
Short description: The number of tests performed.
Latest estimate: 0.67 daily tests per thousand people (as of 14 August 2020).
Detailed description:
The Panama Ministry of Health reports the cumulative number of tests performed (“pruebas realizadas”) on their official dashboard with a time series dating back to 9 March 2020. The page with testing numbers is not the first one shown but can be navigated to with the arrows at the bottom of the dashboard. The dashboard shows the cumulative number of total, positive, negative, and control tests (“prueba de control”) performed. We report here the total of positive and negative numbers because: 1) the time series only includes positive and negative test numbers; and 2) the total they provide seems to include control tests, which we understand to be used for testing quality control.
→ View the country profile of Paraguay for the coronavirus pandemic.
Source: Paraguay Ministry of Public Health and Social Welfare
Short description: The number of tests performed.
Latest estimate: 0.4 daily tests per thousand people (as of 22 August 2020).
Detailed description:
The Paraguay Ministry of Public Health and Social Welfare maintains an official dashboard that reports the weekly and daily number of PCR tests conducted, alongside the percentage of postive tests (‘confirmados’). Weekly figures are provided in the chart titled “Cantidad de pruebas vs confirmados”. Daily figures are available by clicking on the “+” symbol that appears when hovering over the bottom-left corner of this chart.
The fact that the percentage shown equates to the number of confirmed cases suggests that the testing figures equate to the number of people tested and do not include tests pending results.
The reported figures are cumulative from 7 March 2020, when the first case in Paraguay was confirmed.
→ View the country profile of Peru for the coronavirus pandemic.
Source: Peru Ministry of Health
Short description: The number of people tested.
Latest estimate: 0.26 daily tests per thousand people (as of 18 August 2020).
Detailed description:
The Peruvian Ministry of Health publishes daily presentation slides on the number of people tested for the coronavirus, explicitly distinguishing between PCR and rapid tests. We only include PCR test figures in our series.
On dates where presentation slides are unavailable, we use the Ministry’s press releases to supplement the series. Where the dates across the two sources overlap, the number of people tested in the presentation slides are consistently identical to the press release figures for the previous day.
For 22 March 2020, the press release quotes 7486 samples were tested, although adding the total number of positive (395) and negative (6269) tests totals to 6664 samples – consistent with neighbouring test figures. Hence, we use the summed total oof 6664 samples.
Where multiple press releases have been issued in one day, 20 March 2020, we report the testing figures in the most recent release.
→ View the country profile of Philippines for the coronavirus pandemic.
Source: Department of Health
Short description: The number of people tested.
Latest estimate: 0.29 daily tests per thousand people (as of 22 August 2020).
Detailed description:
As part of its COVID-19 dashboard, the Philippines Department of Health provides a ‘COVID-19 DOH Data Drop’ as a series of CSV files shared on Google Drive.
Among these files, the ‘Testing Aggregates’ file includes a time series of cumulative unique individuals tested since 3 April 2020.
The source provides a breakdown by laboratory. Data for recent days may be incomplete due to delays in reporting.
→ View the country profile of Poland for the coronavirus pandemic.
Source #1: Poland Ministry of Health
Short description: The number of people tested.
Latest estimate: 0.6 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Polish Ministry of Health provides a daily diagnostics reports on their official Twitter account that gives the cumulative number of samples tested and people tested.
The earliest observation on figures for people tested is from 28 April 2020.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
Source #2: Poland Ministry of Health
Short description: The number of samples tested.
Latest estimate: 0.63 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Polish Ministry of Health provides a daily diagnostics reports on their official Twitter account that gives the cumulative number of samples tested and people tested.
The earliest observation for the number of samples tested begins from 6 March 2020. Between 12 and 27 April 2020, the number of samples tested is split into the number of positive and negative results. There are some dates for which we could not find the MOH tweets and these gaps are filled in using observations taken from this unoffical repository of the MOH figures.
There is a fall in the cumulative total on 8 August 2020 due to a past mistake has been noticed and the cumulative figure revised on the date to adjust for this.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Portugal for the coronavirus pandemic.
Source: Portugal Ministry of Health dashboard
Short description: The number of samples processed
Latest estimate: 1.31 daily tests per thousand people (as of 12 August 2020).
Detailed description:
The dashboard provides figures for the number of samples processed, both as daily figures and as cumulative figure since 1 March 2020.
Up until 29 April 2020 we had reported a figure of the number of people tested, obtained as the sum of confirmed and unconfirmed cases in the Portugal Ministry of Health (MOH) daily updates. We were alerted by a Technical Advisor within the Cabinet of the Secretary of Health to the fact that these figures only captures people who were reported through the National System of Epidemiological Surveillance which does not include many of the people that get tested but are never entered into the surveillance system (because they do not meet the criteria and go on to test negatively). For this reason we have now switched to the current series. The advisor confirmed to us that this series: includes all the public, private and university labs performing SARS-CoV-2 diagnosis in Portugal; does not include tests pending results; includes only PCR tests.
→ View the country profile of Qatar for the coronavirus pandemic.
Source: Qatar Ministry of Public Health
Short description: The number of people tested.
Latest estimate: 1.9 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Qatar Ministry of Public Health provides daily updates on its website of the cumulative number of people tested to date. It is not clear whether these figures include people for which test results are pending.
The earliest reported figure that we have been able to find is from 14 March 2020, at which point 6,788 tests had been conducted.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Romania for the coronavirus pandemic.
Source: Romanian Government
Short description: The number of tests performed.
Latest estimate: 1.03 daily tests per thousand people (as of 23 August 2020).
Detailed description:
We construct a time series from daily press releases on the Romanian Government website which includes a cumulative figure of the number of tests processed to date (e.g. “Până la această dată, la nivel național, au fost prelucrate 484.782 de teste.”). No other information is provided.
Where unavailable at this main source, some earlier observations were sourced from the press office of the Ministry of Internal Affairs.
Our data for this series is sourced from a non-official repository of official data. As explained in our FAQ here we regularly audit the accuracy of this repository against direct official channels. Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Russia for the coronavirus pandemic.
Source: Government of the Russian Federation
Short description: The number of tests performed.
Latest estimate: 1.87 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The Federal Service for Supervision of Consumer Rights Protection and Human Well-Being publishes a daily report giving the cumulative total of tests performed in Russia. No information is given on the geographical scope and number of labs included.
We do not include cumulative totals reported on 30 March 2020 and 31 March 2020, as they seemed inconsistent with numbers given on 29 March 2020 and 1 April 2020.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Rwanda for the coronavirus pandemic.
Source: Rwanda Ministry of Health
Short description: The number of samples tested.
Latest estimate: 0.38 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The Rwanda Ministry of Health (@RwandaHealth) tweets press releases in English, French, and Kinyarwanda that report the number of samples tested each day.
The earliest press release we have found containing testing figures is from 7 April 2020. Between 7 April 2020 and 1 May 2020 (inclusive), the press releases reported the number of samples tested today, but not the cumulative number of samples tested to date. Since 2 May 2020, the press releases have reported the cumulative number of samples tested to date as well as the number of samples tested today. We construct a daily time series since 6 April 2020 of the cumulative number of samples tested to date by subtracting daily tests between 7 April 2020 and 2 May 2020 from the 2 May 2020 cumulative total.
As of 6 April 2020, 5,701 samples had been tested to date. It is unclear when the first samples were tested.
It is unclear whether the reported figures include tests for which the results are pending.
Prior to 17 April 2020, the press releases reported testing figures using the language “samples tested”. From 17 April 2020 onwards, the press releases have used the ambiguous language “tests today”. We assume that “tests today” still refers to the number of samples tested.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Saudi Arabia for the coronavirus pandemic.
Source: Ministry of Health
Short description: The number of tests performed.
Latest estimate: 1.71 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Saudi Arabian Ministry of Health provides a dashboard detailing the total number of tests performed.
The dashboard reports testing figures as “total tests”, making it unclear whether the figures refer to the number of samples tested, tests performed, or people tested. Nevertheless, corresponding press releases from the Ministry of Health imply that the figures refer to the total number of tests performed. For example, the 26 July 2020 press release states that “…57,216 new Polymerase Chain Reaction (PCR) tests for Novel Coronavirus (COVID-19) have been conducted in the Kingdom’s laboratories, bringing the total number to 3,056,956 lab tests”. This 3,056,956 figure differs slightly from the figure of 3,047,668 tests reported on the official dashboard on 26 July 2020, but we assume this small discrepancy is due to differences in reporting time rather than a substantive difference in units (e.g. tests performed vs. people tested).
The exact date these cumulative figures date back to is unknown.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Senegal for the coronavirus pandemic.
Source: Ministry for Health and Social Action
Short description: The number of tests performed.
Latest estimate: 0.08 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Senegalese Ministry for Health and Social Action publishes daily press releases detailing the number of tests performed (“tests réalisés”) and the number of positive confirmed cases (“cas positifs”). The fact that the number of tests with positive results (“son revenus positifs”) equals the number of confirmed cases implies that the testing figures equal the number of people tested.
The daily press releases date back to 28 February 2020. We construct a daily time series of the number of tests performed using data contained in this unofficial github repository, which we have cross-checked against official data for a sample of dates.
Our data for this series is sourced from a non-official repository of official data. As explained in our FAQ here we regularly audit the accuracy of this repository against direct official channels. Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Serbia for the coronavirus pandemic.
Source: Ministry of Health
Short description: The number of people tested.
Latest estimate: 1.37 daily tests per thousand people (as of 23 August 2020).
Detailed description:
Reports are published daily by the Serbian Ministry of Health. The data is collected and aggregated by volunteers and published on GitHub. All labs in Serbia are included.
Our data for this series is sourced from a non-official repository of official data. As explained in our FAQ here we regularly audit the accuracy of this repository against direct official channels. Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Singapore for the coronavirus pandemic.
Source #1: Ministry of Health
Short description: The number of people tested.
Latest estimate: 1.09 daily tests per thousand people (as of 18 August 2020).
Detailed description:
The Ministry of Health dashboard gives a cumulative total of swabs tested and unique persons tested.
No other information is provided.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
Source #2: Ministry of Health
Short description: The number of swabs tested.
Latest estimate: 3.3 daily tests per thousand people (as of 18 August 2020).
Detailed description:
The Ministry of Health dashboard gives a cumulative total of swabs tested and unique persons tested.
No other information is provided.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Slovakia for the coronavirus pandemic.
Source: National Center of Health Information and the Slovak Republic Government coronavirus information website
Short description: The number of tests performed.
Latest estimate: 0.48 daily tests per thousand people (as of 22 August 2020).
Detailed description:
Slovakia’s National Center of Health Information provides a dashboard with figures for ‘all tests performed’ (“Všetkých vykonaných testov”).
Using web archives, we construct a daily time series of total samples by looking at earlier snapshots of the website. These archives begin from 15 March 2020. We are unable to determine when testing began, only that on 15 March 2020 a total of 1,545 tests had been conducted. Earlier archives make it clear that the figures relate to the sum of positive and negative tests (i.e. tests pending results are not included).
We found the total number of negative and positive samples was identical for 26 March 2020 and 27 March 2020. The reason for this is not clear. We include only the former date as an observation. From 29 March – 1 April 2020 and 5 April – 7 April 2020 no web archives could be retreived.
From 14 April 2020, we became aware of a short-term dashboard by the National Centre of Health and Information providing the same figures over the last 6 days. We use this source to supplement testing data for 10 April 2020 where web archives are not available. The notes to this dashboard provide further elaboration. They state that the figures relate to the “number of completed laboratory tests… The number contains all positive and negative results also retests of previously tested positive patients” (via Google translate).
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Slovenia for the coronavirus pandemic.
Source: Slovenian Government coronavirus information page
Short description: The number of tests performed.
Latest estimate: 0.48 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The Government information website provides figures for the number of tests completed (“Opravljeni testi”). A time series of both cumulative and daily tests is available to download. This dates back to 12 March 2020, where 3,863 tests are reported to have already been performed (it is not known from which date this first cumulative figure dates back to).
The same cumulative figure is published each day by the National Insitute for Public Health (NIJZ). Here they clarify that figures relate to the number of tests – including those for people who were tested several times.
A footnote states that the figures relate to “Laboratory tests performed as part of routine testing and the COVID-19 National Survey are included.” The later appears to refer to the prevalence study described on this Government page. The description of the study states that “The survey is being conducted on a random sample of 3,000 persons” and that people will be tested with both a PCR and a serological test. For this reason, the reported testing figures may include serological tests in addition to PCR tests.
The volunteer-led Sledilnik.org project also presents the official data in a helpful website.
→ View the country profile of South Africa for the coronavirus pandemic.
Source: National Institute for Communicable Diseases (NICD)
Short description: The number of people tested.
Latest estimate: 0.38 daily tests per thousand people (as of 22 August 2020).
Detailed description:
The South African National Institute for Communicable Diseases (NICD) publishes daily updates on the number of confirmed cases, deaths and tests conducted nationally and by province. These updates are published on its website and on its offical Twitter account (@nicd_sa).
The NICD reports the number of ‘tests processed’, which is also labelled as ‘total tested’. On 18 April 2020, the official twitter account for the Department for Health clarified that repeat tests for COVID-19 are not counted and that the testing figures refer to the number of people tested.
The NICD began publishing daily updates on 7 February 2020, allowing us to develop a time series from this date forward. We do not know the first date of testing. As of 7 February 2020, 42 people had been tested.
We source this data from the Data Repository for South Africa repo – created, maintained and hosted by Data Science for Social Impact research group, led by Dr. Vukosi Marivate, at the University of Pretoria.
Our data for this series is sourced from a non-official repository of official data. As explained in our FAQ here we regularly audit the accuracy of this repository against direct official channels. Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of South Korea for the coronavirus pandemic.
Source: South Korea CDC
Short description: The number of people tested.
Latest estimate: 0.25 daily tests per thousand people (as of 24 August 2020).
Detailed description:
KCDC have provided daily updates in English since 21 January 2020.
The daily updates show the change each day and the current total. These form a consistent chain all the way back to 21 January 2020. The daily test figures we provide relate to the daily change in the number of tests with results (the figures do not include those cases pending test results).
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Spain for the coronavirus pandemic.
Source: Ministerio de Sanidad, Consumo y Bienestar Social
Short description: The number of tests performed.
Latest estimate: 1.48 daily tests per thousand people (as of 20 August 2020).
Detailed description:
Official data covering the whole of Spain was only published sporadically until the end of April 2020.
The first estimate in our time series comes from a Ministry of Health press release, quoting Salvador Illa, Minister of Health: “From the beginning of the crisis and until April 13, the total number of PCR carried out in Spain is 930,230.” The press release is here.
The second is another Ministry of Health press release published on 27 April 2020 and relating to testing conducted up to 23 April 2020. It provides a breakdown across Autonomous Communities (regions), with a total for Spain of 1,035,522 PCR tests. The same release provides a figure of 310,038 antibody tests conducted nationwide. These are different to the PCR tests, as we discuss here. Our database aims to not include antibody tests. So in order to provide the data that is most comparable to the other countries in our database, we include only the PCR tests count for this observation.
One aspect of the 27 April 2020 press release that is not totally clear is whether the figures provided for each Autonomous Community relate to the 23 April 2020. Comparing the regional breakdowns provided in the Ministry of Health release to those collected from official sources by CIVIO (as of 28 April 2020) suggests that the figure of 203,892 provided for Madrid in the Ministry of Health update—ostensibly dating to the 23 April 2020—may date back to 14 April 2020.
Further data points are collected from the Ministry of Health’s press releases or its Twitter account.
CIVIO is a non-profit investigative data journalism organisation who are collating information on tests performed within each Autonomous Community, as released through official channels (communication offices, web pages, press releases and social networks of the different regions).
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
—
Daily case reports (e.g. 5 August 2020) published by the Ministry of Health note that “total cases [are] confirmed by PCR until 10 May, and by PCR and IgM (only if compatible symptoms) according to the new surveillance strategy from 11 May”.
This could have indicated that Spain included positive antibody tests in their figures for confirmed cases. As we explain in this FAQ, we do not present figures for the positive rate or number of tests per confirmed case for countries where we believe this to be the case.
However, this footnote importantly says that positive IgM results are only counted if they have “compatible symptoms”, i.e. if patients are actively infected by the virus.
This report published on 9 July 2020 also makes clear on page 4 that IgM tests included in new confirmed cases do not correspond to massive serology testing for past infections, but rather to active, symptomatic infections with a negative PCR result but for which further IgM testing came back positive.
In the same way, the documentation published by the Carlos III Health Institute on its COVID-19 website states that “as of May 11, the Ministry of Health is counting the confirmed cases diagnosed by PCR (and in some specific cases those diagnosed by IgM by ELISA)”, i.e. the inclusion of IgM does not mean the inclusion of massive antibody testing results, but rather specific instances where current infections are confirmed via IgM instead of PCR.
Pending new information that would prove otherwise, we therefore do not exclude Spain from the calculation of the positive rate and number of tests per case, as IgM tests for active cases represent a very small number and therefore should not affect these ratios.
→ View the country profile of Sweden for the coronavirus pandemic.
Source #1: Sweden Public Health Agency
Short description: The number of people tested.
Latest estimate: 1.1 daily tests per thousand people (as of 5 July 2020).
Detailed description:
From the start of the pandemic up until July 2020, the Swedish Public Health Agency provided weekly COVID-19 reports that provide information on the number of people tested.
The first observation in the series is the total people tested from week 4 (calendar week: 20 – 26 January 2020) to week 8 (calendar week: 17 – 23 February 2020).
Source #2: Sweden Public Health Agency
Short description: The number of samples tested.
Latest estimate: 0.8 daily tests per thousand people (as of 16 August 2020).
Detailed description:
Since July 2020, the Public Health Agency has replaced its previous weekly updates of the number of people tested with the weekly count of samples tested. This data does not allow us to publish any cumulative total, since no data is available on how many samples had been tested prior to the first week of reporting. However, we integrate this data into our “daily tests” time series by dividing each weekly total by 7, thereby estimating the average daily number of samples tested.
→ View the country profile of Switzerland for the coronavirus pandemic.
Source: Federal Office of Public Health
Short description: The number of tests perfomed.
Latest estimate: 0.92 daily tests per thousand people (as of 22 August 2020).
Detailed description:
The Federal Office of Public Health presents a time series of daily positive and negative tests, downloadable as a csv file on their website. A note states that: “The data published here is based on information that laboratories, doctors and hospitals have given us. They refer to reports that we received this morning and can therefore deviate from the figures that the cantons communicate.” (via Google translate).
At this page they also make available daily situation reports, which present the same figures, making it explicit that the numbers refer to PCR tests.
The data can also be accessed by downloading the graphic software file in their visualization here. This graphic notes that “Since several tests can be taken and reported per person, the number of positive tests is higher than the number of positively tested people”. (via Google translate).
→ View the country profile of Taiwan for the coronavirus pandemic.
Source: Taiwan CDC Open Data Portal
Short description: The number of people tested.
Latest estimate: 0.01 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Taiwan Centers for Disease Control (CDC) publishes a complete time series of the number of cases tested on its open data portal.
A technical specialist from the Taiwan CDC confirmed to Our World in Data that “in Taiwan, every notified case tests for SARS-CoV-2. The number of tests performed per notified case relate to the criteria for releasing COVID-19 confirmed cases from isolation, and the disease status of notified cases. There are three reporting portals provided for medical workers and public health workers to report a COVID-19 suspected case: “法定傳染病通報”, “居家檢疫送驗”, “擴大監測送驗”. The number of COVID-19 related tests is the sum of column B(法定傳染病通報), C(居家檢疫送驗), and D(擴大監測送驗).”
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Thailand for the coronavirus pandemic.
Source #1: Department of Disease Control
Short description: The number of people tested.
Latest estimate: 0.03 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Thailand Department of Disease Control (DDC) issues daily situation reports which detail the status of confirmed cases, deaths and people being assessed.
We provide the figures reported as “persons under investigation” (PUI). PUI is the sum of confirmed cases, the number of people who tested negative, and the number of people for whom results are still pending. All three figures are reported individually from 3 March until 31 March 2020. From 1 April 2020 onwards, the number of people who tested negative and the number for whom results are still pending are no longer reported separately.
We have received communication from Thailand’s Ministry of Public Health (MOPH) that individuals who did not meet PUI criteria can pay out of pocket to have their samples tested, referred to as the “non-PUI” group. This “non-PUI” group is large in the context of Thailand. This “non-PUI” group is not reflected in our current figures.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
Source #2: Department of Disease Control
Short description: The number of tests performed.
Latest estimate: 0.03 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Thailand Department of Disease Control (DDC) issues daily situation reports which detail the status of confirmed cases, deaths and people being assessed.
We provide the figures reported as “number of laboratory examinations”. The reported figures include PCR tests only. It is unclear whether pending test results are included.
Prior to 1 July 2020, we collected data from the World Health Organization Country Office for Thailand, which provides COVID-19 situation reports that periodically report the cumulative number of samples tested to date. The reported figures include samples tested in both public and private labs. However, reporting delays appear to affect the figures. For example, the figures for 1 May 2020 reflect only 121 of 142 public and private laboratories certified for PCR testing. The earliest figure we have found is for 10 April 2020, at which point 100,498 samples had been tested. The reported figures are cumulative, but the situation reports do not state an exact reference date.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Togo for the coronavirus pandemic.
Source: Togo Ministry of Health
Short description: The number of samples tested.
Latest estimate: 0.08 daily tests per thousand people (as of 19 August 2020).
Detailed description:
The Togo Ministry of Health provides daily updates on the number of samples tested via their official government portal and affiliated social media accounts on Facebook and Twitter.
We are in direct contact with the MOH who provide daily updates on the number of samples tested. These figures are very similar to the sources listed above but differs a little due to internal MOH reporting delays. Our figures are more up-to-date. For example, for the 28 June 2020 the cumulative total of samples tested on Twitter is reported to be 30,316 samples while in the internal MOH figures it is 30,333. Please see the notes for each observation where we include a link to the corresponding MOH tweet for comparison, where available.
The earliest observation is available from 4 March 2020.
The figures refer to the number of PCR tests performed each day.
→ View the country profile of Tunisia for the coronavirus pandemic.
Source: Tunisian Ministry of Health
Short description: The number of tests performed.
Latest estimate: 0.15 daily tests per thousand people (as of 21 August 2020).
Detailed description:
The Tunisian Ministry of Health dashboard provides daily figures of the total number of cases tested (‘Nombre de cas journaliers’). They also provide a time series for the number of tests per day (‘Nombre de tests journaliers’). The figures we provide relate to the latter.
The earliest observation is available from 10 March 2020. However, it is unclear whether this is also when testing first began.
The MOH does not report the number of tests performed on 5 May 2020.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Turkey for the coronavirus pandemic.
Source: Turkish Ministry of Health
Short description: The number of tests performed.
Latest estimate: 1.06 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Turkish Ministry of Health now publishes a daily chart of Coronavirus confirmed cases, deaths, recoveries, total tests (“TOPLAM TEST SAYISI”), and tests today (“BUGÜNKÜ TEST SAYISI”).
These daily charts are published on the Turkish Ministry of Health’s website – previous versions can be found on Web Archive. This only extends back to 27 March 2020. Prior to this date, we rely on figures reported by the Turkish Minister for Health at daily press conferences. In these reports they detail the latest update of confirmed cases, deaths and tests conducted.
We are not aware of the date that testing began: only that as of 18 March 2020 – the first data point in our series – 10,018 tests had been conducted.
No other information concerning the figures is known.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Uganda for the coronavirus pandemic.
Source: Uganda Ministry of Health
Short description: The number of samples tested.
Latest estimate: 0.08 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The Uganda Ministry of Health publishes a daily press release detailing the number of cumulative and daily samples tested. As a result of changes in the way the data is published, we only report the cumulative total of samples tested when it is published by the official source.
From late June 2020, the press releases regularly include the official cumulative number of samples tested. Prior to this date, only daily testing figures were reported – with the exception of reports from 6–14 April 2020.
The earliest press release we could find that lists the cumulative and daily figures is for 1 April 2020. However, we cannot say with certainty when testing began and the precise date from which cumulative totals begin.
→ View the country profile of Ukraine for the coronavirus pandemic.
Source: Cabinet of Ministers of Ukraine
Short description: The number of tests performed
Latest estimate: 0.41 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The Cabinet of Ministers of Ukraine provides daily snapshots that report the cumulative number of tests performed to date.
It is unclear whether the reported figures include tests pending results.
In the daily snapshots, the wording used to report the cumulative testing figures does not make it clear as to whether the reported figures relate to the number of tests performed or the number of persons tested. For example, the cumulative testing figure reported on 6 August 2020 is “1,116,641 tested” (translated from “протестовано”). The english version of the same page reports the same figure as “total of tests”, which is also unclear. A more detailed dashboard on the same web domain provides testing figures in downloadable csv files dating back to 1 June 2020, where each csv file contains separate columns for the cumulative number of PCR tests and non-PCR tests. The cumulative PCR testing figures reported in these csv files are described as “The number of laboratory tests performed by PCR on COVID-19” (“Кількість проведених лабораторних досліджень методом ПЛР на COVID-19”), making it clear that the figures refer to the number of tests performed (which may or may not be equivalent to the number of people tested).
The cumulative PCR testing figures provided in the csv files tend to be approximately 2% smaller than the testing figures provided in the daily snapshots. Nevertheless, we assume that this discrepancy is due to small differences in the reference date, reporting laboratories included, or other minor details, rather than the possibility that the daily snapshot figures include non-PCR tests or refer to a metric other than the number of tests performed. The cumulative number of non-PCR tests reported in the csv files are much larger than the size of this discrepancy.
The earliest reported figure that we have been able to find is from 9 April 2020, at which point 20,608 cumulative tests were reported.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of United Arab Emirates for the coronavirus pandemic.
Source: UAE Federal Competitiveness and Statistics Authority
Short description: The number of tests performed.
Latest estimate: 7.2 daily tests per thousand people (as of 24 August 2020).
Detailed description:
The UAE Federal Competitiveness and Statistics Authority provides a time series dataset (available to download), including the number of daily tests conducted from 29 January 2020 to today.
The source notes “To ensure the highest possible quality and accuracy of the COVID-19 numbers, these data are regularly reviewed by UAE governmental authorities concerned. Numbers may change as the data are continuously refined.”
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of United Kingdom for the coronavirus pandemic.
Source: Public Health England/Department of Health and Social Care
Short description: The number of tests performed.
Latest estimate: 2.39 daily tests per thousand people (as of 19 August 2020).
Detailed description:
Since early June 2020, the United Kingdom has been publishing a full retrospective time series in CSV format, with testing data going back to March 2020. We aggregate the data for the United Kingdom based only on two ‘pillars’:
– Pillar 1: swab (antigen) testing in Public Health England (PHE) labs and NHS hospitals for those with a clinical need and health and care workers;
– Pillar 2: swab (antigen) testing for the wider population.
For both pillars, we use the ‘Daily number of tests processed’ and ‘Cumulative number of tests processed’ columns to create of time series of PCR tests performed in the United Kingdom. We do not include tests delivered or sent out that have not been recorded as having been processed.
We do not include other pillars, i.e.:
– Pillar 3: this pillar consists (fully) of serology tests, which we aim not to include;
– Pillar 4: this pillar consists (partly) of serology tests, which we aim not to include, and positive tests results from this pillar seem not to be included in the government’s total for positive cases.
The source notes that ‘data on UK tests is updated daily. The figures for test results are compiled from different sources. Daily totals reflect actual counts reported for the previous day. Each day there may be corrections to previous reported figures. This means that previously published daily counts will not necessarily sum to the latest cumulative figure. It also means that today’s cumulative count may not match the previous day’s cumulative count plus today’s daily count.’
→ View the country profile of United States for the coronavirus pandemic.
Source #1: COVID Tracking Project
Short description: The data are an aggregation of figures released by individual states. The Project aim to report on the number of people tested, including private labs, but not all states report their figures in this way. The figures for some states appear to include antibody tests in addition to PCR tests.
Latest estimate: 2.04 daily tests per thousand people (as of 23 August 2020).
Detailed description:
This is a collaborative project launched in order to fill some of the important gaps in the testing figures being collated by the CDC.
Testing data is gathered from individual states, as reported in state health department websites, data dashboards and press releases from officials.
As of 26 May 2020, there have been a number of media reports noting that the testing figures released by some states include antibody tests in addition to PCR tests – as discussed in this article in The Atlantic. Our dataset aims to report only PCR tests. But because some states do not disaggregate these types of tests we are not currently able to exclude the antibody tests.
Other differences across states include: some report the number of tests performed, others the number of people tested; some include private labs, others not; some report negative test results, others only positive test results; some include pending tests, others do not (below we show figures that exclude explicitly pending results).
Moreover, many states do not explicitly provide details about these important factors needed to interpret the data they provide.
There are issues in comparing the figures over time. The totals given for early on in the outbreak do not include all states. One significant uncertainty is the extent to which the rapid rise in tests seen from the mid-March in part reflects states beginning to report private lab tests.
Overall the figures should be seen as providing a general indication of testing output, rather than a specific count of a given indicator. Given the very incomplete coverage and reporting delays of the CDC data, it provides a very important additional perspective.
The Project documents their work in lots of detail. See the link provided above for full details.
The United States testing data includes non-PCR tests, which reduces its comparability with other countries.
Source #2: United States CDC
Short description: The number of tests performed. The figures are the sum across states, some of which appear to include antibody tests in addition to PCR tests.
Latest estimate: 2.69 daily tests per thousand people (as of 19 August 2020).
Detailed description:
On 13 May 2020, the United States CDC’s testing data was modified to include tests performed by commercial, hospital, public health, and CDC laboratories. This new data is available on the Testing Data in the US page and CDC COVID Data Tracker.
Initially, this data was explicitly described by the CDC as only including viral tests and excluding antibody tests. As recently as 18 May the site stated that ‘These data only represent viral tests. Antibody tests are not currently captured in these data.’
However, as of 26 May 2020, there have been a number of media reports noting that the testing figures released by some states – which are aggregated by the CDC – include antibody tests in addition to PCR tests, as discussed in this article in The Atlantic.
Within this time, the CDC have also changed the way it described the data to be ambiguous as to whether anitbody tests may be included in their figures (see 26 May 2020 update). Our dataset aims to report only PCR tests. But because any antibody tests in the CDC data are not disaggregated, we are unable to exclude them.
No time series is published for this new version of the data. We therefore started collecting on 13 May 2020 the cumulative total of viral tests reported, which is updated daily.
As of 26 May 2020, the notes to the data provided by the CDC are limited. They state:
– These data are compiled from a number of sources. Not all tests are reported to CDC.
– The number of positive tests in a state is not equal to the number of cases, as one person may be tested more than once.
The CDC previously published a time series that only covered public health labs and did not include private lab tests, which were occurring in significant numbers. Daily figures were provided since 18 January. This data is still visible on this page of the CDC website.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
The United States testing data includes non-PCR tests, which reduces its comparability with other countries.
→ View the country profile of Uruguay for the coronavirus pandemic.
Source: Ministry of Public Health
Short description: The number of tests performed.
Latest estimate: 0.57 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The Ministry of Public Health in Uruguay publishes daily reports providing the cumulative total of tests performed. These figures are also made available on the Sistema Nacional de Emergencias dashboard, in which the figures are presented as the number of tests processed since 13 March 2020 (“Desde el 13 de marzo se han procesado: [N] test”).
We previously reported testing data from a dashboard maintained by El Observador, but on 20 April 2020 we replaced the entire time series based on the reports from the Ministry of Health.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Vietnam for the coronavirus pandemic.
Source: Vietnamese Ministry of Health
Short description: The number of samples tested.
Latest estimate: 0.11 daily tests per thousand people (as of 29 April 2020).
Detailed description:
Until the end of April 2020, the Vietnamese Ministry of Health were updating a figure for tests on its disease situation statistics page daily.
The dashboard referred to its testing data as ‘samples tested’. This figure generally matches the sum of confirmed cases and negative samples, suggesting that the total number of samples tested may be equivalent to the number of people tested (there are occasionally small discrepancies – we report the sum of positive and negative in these cases). However, it is unclear whether ‘negative samples’ includes instances of resampling for the same person.
Using web archives we can reconstruct a daily time-series of total samples tested. These daily updates extend back to 2 March 2020. We cannot say with certainty when testing began, only that 1,753 samples had been tested as of 2 March 2020. More recently the website stopped updating its testing data, and finally stopped reporting the figures altogether. The last date we were able to collect testing data was on the 29 April 2020.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
→ View the country profile of Zimbabwe for the coronavirus pandemic.
Source: Zimbabwe Ministry of Health and Child Care
Short description: The number of tests performed.
Latest estimate: 0.08 daily tests per thousand people (as of 23 August 2020).
Detailed description:
The Zimbabwe Ministry of Health and Child Care provides daily press releases on its website and Twitter account (@MoHCCZim) that report the cumulative number of tests performed to date. The reported figures include positive, negative, and pending test results.
The press release for 6 May 2020 reported that 7,808 PCR and 8,244 “rapid screening” tests had been conducted to date. We exclude rapid screening tests from the daily time series that we construct, since we assume that these are antibody tests.
Prior to 6 May 2020, the press releases either: (a) reported a combined cumulative total of PCR and rapid screening tests without providing a breakdown between the two types of tests; or (b) did not clearly specify whether the reported cumulative total was in reference to PCR tests, antibody tests, or both. For this reason, the daily time series we construct begins on 6 May 2020, at which point the press releases began to clearly indicate that the reported cumulative totals only include PCR tests.
Note that, due to the way the data is presented by the official source, the time series may be impacted by retrospective revisions made by the source – see our FAQ here.
Other countries
Below is the list of countries for which we have attempted to collect data but could not find official sources.
- Afghanistan (last checked on 23 June 2020): no data from official sources could be found.
- Algeria (last checked on 05 August 2020): no data from official sources could be found.
- American Samoa (last checked on 08 May 2020): no data from official sources could be found.
- Andorra (last checked on 08 May 2020): no data from official sources could be found.
- Angola (last checked on 04 August 2020): no data from official sources could be found.
- Aruba (last checked on 08 June 2020): no data from official sources could be found.
- Benin (last checked on 08 June 2020): no data from official sources could be found.
- British Virgin Islands (last checked on 08 June 2020): no data from official sources could be found.
- Burkina Faso (last checked on 10 June 2020): NA
- Burundi (last checked on 08 June 2020): no data from official sources could be found.
- Cambodia (last checked on 08 June 2020): no data from official sources could be found.
- Central African Republic (last checked on 09 June 2020): no data from official sources could be found.
- Chad (last checked on 09 June 2020): no data from official sources could be found.
- China (last checked on 09 June 2020): we are not aware of any figures relating to the whole of China. There is some data relating to parts of China, for instance press releases concerning testing in Guangdong province.
- Comoros (last checked on 09 June 2020): no data from official sources could be found.
- Congo (last checked on 09 June 2020): no data from official sources could be found.
- Curacao (last checked on 09 June 2020): no data from official sources could be found.
- Equatorial Guinea (last checked on 09 June 2020): no data from official sources could be found.
- French Guiana (last checked on 09 June 2020): no data from official sources could be found.
- Gabon (last checked on 09 June 2020): no data from official sources could be found.
- Georgia (last checked on 10 June 2020): no data from official sources could be found.
- Guam (last checked on 10 June 2020): no data from official sources could be found.
- Guatemala (last checked on 10 June 2020): no testing data from official sources could be found.
- Guinea (last checked on 10 June 2020): no data from official sources could be found.
- Guinea-Bissau (last checked on 10 June 2020): no data from official sources could be found.
- Haiti (last checked on 10 June 2020): no data from official sources could be found.
- Kiribati (last checked on 10 June 2020): no data from official sources could be found.
- Lesotho (last checked on 11 June 2020): no data from official sources could be found.
- Liberia (last checked on 11 June 2020): no data from official sources could be found.
- Libya (last checked on 11 June 2020): no data from official sources could be found.
- Liechtenstein (last checked on 11 June 2020): no data from official sources could be found.
- Macao (last checked on 11 June 2020): no data from official sources could be found.
- Marshall Islands (last checked on 11 June 2020): no data from official sources could be found.
- Mauritius (last checked on 12 June 2020): no data from official sources could be found.
- Micronesia (country) (last checked on 12 June 2020): no data from official sources could be found.
- Monaco (last checked on 12 June 2020): no data from official sources could be found.
- Mongolia (last checked on 12 June 2020): no data from official sources could be found.
- Montenegro (last checked on 12 June 2020): no data from official sources could be found.
- Nauru (last checked on 12 June 2020): no data from official sources could be found.
- Nicaragua (last checked on 12 June 2020): no data from official sources could be found.
- Niger (last checked on 12 June 2020): no data from official sources could be found.
- North Korea (last checked on 12 June 2020): no data from official sources could be found.
- Northern Mariana Islands (last checked on 12 June 2020): no data from official sources could be found.
- Palau (last checked on 12 June 2020): no data from official sources could be found.
- Papua New Guinea (last checked on 13 June 2020): no data from official sources could be found.
- Puerto Rico (last checked on 13 June 2020): no data from official sources could be found.
- Saint Martin (French part) (last checked on 13 June 2020): no data from official sources could be found.
- Samoa (last checked on 14 June 2020): no data from official sources could be found.
- Sao Tome and Principe (last checked on 14 June 2020): no data from official sources could be found.
- Seychelles (last checked on 14 June 2020): no data from official sources could be found.
- Solomon Islands (last checked on 14 June 2020): no data from official sources could be found.
- Somalia (last checked on 14 June 2020): no data from official sources could be found.
- South Sudan (last checked on 14 June 2020): no data from official sources could be found.
- Sudan (last checked on 23 July 2020): NA
- Suriname (last checked on 14 June 2020): no data from official sources could be found.
- Syrian Arab Republic (last checked on 14 June 2020): no data from official sources could be found.
- Tajikistan (last checked on 14 June 2020): no data from official sources could be found.
- Tanzania (last checked on 05 August 2020): The Tanzania Ministry of Health press releases report the number of confirmed cases, but not the number of tests conducted. We have been unable to find official testing figures from any other official source.
- Tonga (last checked on 14 June 2020): no data from official sources could be found.
- Turkmenistan (last checked on 14 June 2020): no data from official sources could be found.
- Tuvalu (last checked on 14 June 2020): no data from official sources could be found.
- Uzbekistan (last checked on 14 June 2020): no data from official sources could be found.
- Venezuela (last checked on 14 June 2020): no testing data from official sources could be found.
- Yemen (last checked on 14 June 2020): no data from official sources could be found.
Frequently Asked Questions about our COVID-19 Testing dataset
We do not have testing data for all countries.
For some countries we are aware of a source of data and are currently in the process of adding it to our dataset.
For others, we are not aware of any official source of testing data.
If you are aware of a source of testing data for a country that we do not currently list, you can help us by filling out this Google Form.
Testing data is collected by Our World in Data by browsing public information from official sources. We rely on figures published on official websites, in press releases and by social media accounts of national authorities—usually governments, ministries of health, or centres for disease control.
Our testing dataset is entirely replicable:
- Links to the main sources for each country are provided in detailed source descriptions;
- We list specific individual sources for each data point in files available on GitHub.
We regularly update our testing data—around twice a week. You can see when we last updated the dataset at the top of this section.
We only use information made publicly available by each country. Many countries do not release figures on a daily basis.
Few countries explicitly state whether their published figures include tests pending results or not. We note any available information about this in the data descriptions for each country.
Where figures including and excluding pending tests are both available, we exclude them from our figures. The details of this are always documented in the data descriptions for each country.
As a minimum requirement, we only include figures in the dataset that count both positive and negative test outcomes (and if included, pending results too) according to a consistent unit (i.e. the figures for both positive and negative test outcomes refer either to the number of tests, or the number of people tested, but not a mix of the two units).
Anyone can track changes to our data and updates using our GitHub repository. Each file presents a full history of changes (commits). This allows viewers to trace back any addition or edit to a particular date and time.
Our World in Data does not report data privately to any party. Similarly, countries do not report privately to Our World in Data either.
Every source we consult is public, our data collection can be replicated, and our numbers can be used by anyone. Any person or organization can consult and use our database.
In particular, Our World in Data does not report directly to international organizations. When these organizations want to rely on our data, they visit our website, and access the same public information that is available to everyone.
Our World in Data highlights ambiguities or problems with official sources; but this process is entirely done based on public information, and there is no direct verification with government officials. We list all ambiguities and data problems for each country in the detailed source descriptions.
Where countries provide disaggregated testing figures for different testing technologies, we will include only PCR tests in this dataset. The reason for this is explained in our background information on testing.
Few countries currently explicitly state the testing technology to which their figures relate. However, many countries refer to their testing figures as relating to ‘laboratory’, ‘diagnostic’, or ‘swab’ testing. In general, this is an indication that figures relate to PCR tests, particularly in the context that this technology remains the WHO recommended basis for case confirmation.9
As of 4 May, we are only aware of official data on antibody tests published by Spain and Ecuador.
Our data on testing aims to include only PCR tests for COVID-19 infection. One of the reasons we do this is because in most countries this testing technology is the basis of case confirmation. Positive antibody tests are typically not included in confirmed case counts. As such, it is the number of PCR tests that is the appropriate comparison with confirmed cases when calculating the share of tests that are positive.
Some countries, however, do include positive antibody tests in their figures for confirmed cases. Our testing figures – which exclude antibody tests – are not an appropriate comparison in these instances: on this basis there could be more cases than tests, which is not possible.
For this reason, we do not calculate the positive rate or number of tests per confirmed case for countries where we are aware of antibody tests being used for case confirmation. We also do not show them on scatter charts comparing the number of tests and confirmed cases.
As of 6 August 2020, this includes: Peru, Ecuador and Brazil.
We do plan to publish a new dataset showing figures specifically on antibody tests. This is an area where we plan to expand as more countries adopt systematic antibody testing strategies.
For a small number of countries where data collection from the original official source is prohibitively difficult, we rely on non-official repositories of this official data.
To monitor the ongoing reliability of these non-official repositories, we employ an audit process. Each day three observations are randomly drawn out of all observations in the dataset that have been obtained via such sources. For each selected observation, the recorded figure is manually checked against the direct official channel from which the repository purports to obtain the data. The sampling rate means that each third-party source we make use of is checked around once a week. Given that any discrepancies with official channels are likely to be clustered within particular sources, this provides a high degree of quality control on these sources on a timely basis.
Where any discrepancies are noticed, we switch sources (for the entire time series) to either a different repository or to manual data collection directly from the official channel.
Due to the efforts to produce timely data, official testing figures are subject to frequent retrospective revisions. This can occur for instance where some laboratories have longer reporting delays than others, and previously uncounted tests are then subsequently included.
A number of the sources our database relies on provide a full updated time series. In these cases, such retrospective revisions pose no problem: our full series will reflect the source’s latest information for how many tests occurred on each day.10
However, many of the sources we rely on only provide a daily ‘snapshot’ of the current total figure. In these cases, such retrospective revisions are reflected in our time series according to the day the revision is made, and not when the revised tests actually occurred. This is a distortion of the true trend in testing.
In particular, the figures for daily new tests that we derive from such sources can appear more volatile that they are in reality. In extreme cases, where retrospective revisions are large enough to cause a fall in the reported cumulative number of tests from one day to the next, our series will show a negative number of daily new tests.
For this reason, the rolling average generally provides a more reliable picture than raw daily testing figures.
We would like to acknowledge and thank a number of people in the development of this work: Carl Bergstrom, Bernadeta Dadonaite, Natalie Dean, Jason Hendry, Adam Kucharski, Moritz Kraemer and Eric Topol for their very helpful and detailed comments and suggestions on earlier versions of this work. Tom Chivers we would like to thank for his editorial review and feedback.
And we would like to thank the many hundreds of readers who give us feedback on this work every day. Your feedback is what allows us to continuously clarify and improve it. We very much appreciate you taking the time to write. We cannot respond to every message we receive, but we do read all feedback and aim to take the many helpful ideas into account. Thank you all.