Dear CoronaWatchers,

One year after the start of the CoronaWatchNL project, the coronavirus is still with us. As a community, we made an extensive collection of data on COVID-19 case counts Findable, Accessible, Interoperable, and Reusable (FAIR) (Wilkinson, M. D. et al., 2016). During the first wave of the COVID-19 outbreak in The Netherlands, our project was the primary source of structured open data for researchers, hospitals, (local) governments, and the public. In June 2020, RIVM started to publish their first open and structured data on COVID-19 case counts. More and more users of our project migrated to the RIVM open data.

It's essential to have data openly available and FAIR. Data is an important building block for nowadays research and policymaking. We see that many researchers and organizations struggle with making data and software FAIR. It is important to realize that FAIRness of data is a step-by-step process, and there is no such a thing as a perfect FAIR dataset. We see that many organizations with an important role in the COVID-19 pandemic in the Netherlands started to publish their data openly and take their first steps in making data FAIR. It is still far from perfect, but we are moving forward.

For the award-winning CoronaWatchNL project, this implies we are no longer the connecting link between the user and the suppliers of the COVID-19 data. Users can now make use of the data of RIVM, LCPS, and NICE directly. We have become largely redundant, and therefore we decided to no longer update the project. Our main goal was to become redundant, so we are pleased with the outcome. We will keep an eye on the developments and stay in contact with the suppliers of COVID-19 data. The journey for them has only yet begun.

Users who are still using our data should migrate. Our main sources of data in CoronaWatchNL were RIVM, LCPS, and NICE. Most of the datasets we offer are nowadays available on their websites. See the following sources for more information:

• RIVM: https://data.rivm.nl
• LCPS: https://lcps.nu/datafeed/
• NICE: https://www.stichting-nice.nl/

Now we are no longer updating the data anymore; I would like to thank the CoronaWatchNL community and users. This project was an open community project from the start. Without the help of more than 50 CoronaWatchers, it wouldn't have been possible to collect this amount of data for more than a year. The importance and quality of the data collection were widely recognized in academia, and we were awarded the Dutch Data Prize 2020. Hopefully, we see each other in the future in a new project!

Feel free to contact me with questions: j.debruin1@uu.nl.

Best regards,

Jonathan de Bruin

Dataset: COVID-19 case counts in The Netherlands

CoronaWatchNL collects numbers on COVID-19 disease count cases in The Netherlands. The numbers are collected from various sources on a daily basis, like RIVM (National Institute for Public Health and the Environment), LCPS (Landelijk Coördinatiecentrum Patiënten Spreiding), NICE (Nationale Intesive Care Evaluatie), and the National Corona Dashboard. This project standardizes, and publishes data and makes it Findable, Accessible, Interoperable, and Reusable (FAIR). We aim to collect a complete time series and prepare a dataset for reproducible analysis and academic use.

Dutch:

CoronalWatchNL verzamelt ziektecijfers over COVID-19 in Nederland. Dagelijks worden de cijfers verzameld van het RIVM (Rijksinstituut voor de Volksgezondheid en Milieu), LCPS (Landelijk Coördinatiecentrum Patiënten Spreiding), NICE (Nationale Intesive Care Evaluatie) en Nationale Corona Dashboard. Dit project standaardiseert en publiceert de gegevens en maakt ze vindbaar, toegankelijk, interoperabel en herbruikbaar (FAIR). We streven ernaar om een dataset beschikbaar te stellen voor reproduceerbare analyses en wetenschappelijk gebruik.

Datasets

The datasets available on CoronaWatchNL are updated on a daily basis. Availability depends on the publication by the respective sources (N.B. since July 1st, the epidemiological reports published by RIVM will be released on a weekly instead of a daily basis). The CoronaWatchNL project divides the datasets into four main categories:

• Geographical data
• Descriptive data
• Intensive care data
• Dashboard data
• Miscellaneous datasets

For (interactive) applications based on these datasets, have a look at the applications folder. For predictive models based on these datasets, check out the parallel repository CoronaWatchNL Extended. Please note that the intention of these (too) simplistic models - made by CoronaWatchNL volunteers - is to show how the data can be used for modelling, not to answer specific hypotheses or follow scientific protocol.

Please see the Remarks document for notes about the datasets. Do you have remarks? Please let us know.

Geographical datasets

Reference time: 10:00 AM

These datasets describe the new and cumulative number of confirmed, hospitalized, and deceased COVID-19 cases. Every day, RIVM retrieves the data from the central database OSIRIS at 10:00 AM. Here, the datasets are categorized by their geographical level (i.e., national, provincial, municipal).

For more detail about the specific structure of these geographical datasets, have a look at the data-geocodebook.

Reference time: by day (0:00 AM)

These datasets describe the new and cumulative number of confirmed, hospitalized, and deceased COVID-19 cases per day. The data is retrieved from the central database OSIRIS and counts the number of cases per day (0:00 AM) by RIVM. The dataset concerns numbers on a national level.

For more detail about the specific structure of this geographical dataset, have a look at the data-geocodebook.

Visualizations geographical data

To get a better picture of the content of the geographical datasets, have a look at the following visuals. These visuals show the development of the COVID-19 disease outbreak on a national level.

Descriptive datasets

The datasets in this section describe the new and cumulative number of confirmed, hospitalized, and deceased COVID-19 cases per day and contain variables like age and sex.

For more detail about the specific structure of these descriptive datasets, have a look at the data-desccodebook.

Visualizations descriptive data

The graphs below visualize the development of the COVID-19 disease outbreak per sex and age group.

Intensive care datasets

The intensive care datasets describe the new and cumulative number of COVID-19 intensive care unit (ICU) admissions per day. The datasets are categorized by their source. Compared to RIVM (reporting COVID-19 hospital admissions), CoronaWatchNL collects COVID-19 related intensive care data from LCPS and NICE.

• RIVM reports hospitalized COVID-19 cases, including - but not limited to - the intensive care unit (ICU) admissions. These are the largest numbers and most inclusive counts.
• NICE only reports COVID-19 cases that are admitted to the ICU.
• LCPS, similarly to NICE, reports COVID-19 ICU admissions. However, LCPS tries to compensate for the reporting lag, by estimating its size and adding it to the numbers reported by NICE.

For more detail about the specific structure of the intensive care datasets, have a look at the data-iccodebook.

Visualizations intensive care

The first two graphs show the number of new (Nieuw), total (Actueel), cumulative (Cumulatief), deceased (Overleden), and survived (Overleefd) COVID-19 ICU admissions per day, as declared by NICE. The total number of ICU admissions per day as reported by LCPS is also shown.

Dashboard datasets

The datasets underlying the National Dashboard are listed in this folder. These datasets concern various topics, such as an overview of the number and age distribution of hospitalized, positively tested, and suspected cases, an estimate of the number of contagious people, the reproduction index, the number of (deceased) infected nursery home residents, and the amount of virus particles measured in the sewage water.

For more detail about the specific structure of the dashboard datasets, have a look at the data-dashboardcodebook.

Visualizations dashboard data

These visuals show the development of the COVID-19 disease outbreak on a national level as reported by the National Dashboard and by the RIVM reports.

Below, the number of suspected COVID-19 patients as registered by the GPs, and the amount of COVID-19 particles per milliliter sewage water are depicted.

The reproduction index and estimated contagious people are plotted with their corresponding minimum and maximum values. The reproduction index indicates how quickly the COVID-19 virus is spreading in the Netherlands. The estimated contagious people represent the number of COVID-19 people per 100.000 inhabitants that are contagious for others.

The number of (deceased) nursery home residents infected with COVID-19 are shown here.

Miscellaneous datasets

This folder contains datasets describing various miscellaneous topics, such as the number of (positively) tested people, the underlying conditions and/or pregnancy of deceased cases younger than 70, an overview of the reinforced measures and press releases in the Netherlands, and a list of companies that requested and received an advance on their reimbursement.

For more detail about the specific structure of the miscellaneous datasets, have a look at the data-misccodebook.

Visualizations miscellaneous data

These graphs display the number of (positively) tested people per week. The end date of each week - Sunday - is used as indicator for the respective week.

Below, the cumulative number of deceased COVID-19 cases younger than 70 with and without underlying conditions and/or pregnancy are displayed per notification date.

The cumulative number of specific conditions found in these deceased COVID-19 cases are shown here.

Inactive/deprecated datasets

Deprecated (pending)

The following datasets are awaiting deprecation. They are (being) replaced by new datasets.

* This dataset is extracted from the maps in the PDF's. The values are relative counts per 100.000 residents in the municipality.

Inactive

The following datasets are no longer appended with new data (because RIVM is no longer providing the data).

* Nowadays, the data is published again. Please use dataset data-geo#municipal.

Raw data

CoronaWatchNL collects copies of the raw data such that data collection is verifiable. Copies of the collected data can be found in the folder raw_data. The data isn't standardised.

Data collection sources

The following sources are used for data collection.

License and academic use

The graphs and data are licensed CC0. The original data is copyright RIVM.

For academic use, use presistent data from . This is a persistent copy of the data. Version number refer to the date. Please cite:

De Bruin, J. (2020). Number of diagnoses with coronavirus disease (COVID-19) in The Netherlands (Version v2020.3.15) [Data set]. Zenodo. http://doi.org/10.5281/zenodo.3711575

Image from iXimus via Pixabay

CoronaWatchNL

CoronaWatchNL is collective of researchers and volunteers in The Netherlands. We aim to make the reported number on COVID-19 disease in The Netherlands FAIR. The project is initiated and maintained by Utrecht University Research Data Management Support and receives support from Utrecht University Applied Data Science.

Help on this project is appreciated. We are looking for new datasets, data updates, graphs and maps. Please report issues in the Issue Tracker. Want to contribute? Please check out the help wanted tag in the Issue Tracker. Do you wish to share an application based on these datasets? Have a look at the applications folder. For predictive models, check out the parallel repository CoronaWatchNL Extended.

Please send an email to jonathandebruinos@gmail.com and/or r.voorvaart@uu.nl