Promoting
100

Bewertung von Kontrollmassnahmen

Verlässliche Daten über Massnahmen zur bekämpfung der Epidemie sammeln und ihre Wirksamkeit bewerten.


Evaluation of control measures

Challenge: Collect reliable data on epidemic control measures and assess their effectiveness.

⚙ Repository ✍️ Notebook ►Presentation


The federal council has announced the temporary closure of schools and public places non essential to daily life, ordered the cancellation of mass events and banned the gathering of more than 5 persons. These measures follow -and partially replicate at a federal level- similar measures taken earlier by local authorities and other organisations. These strict measures are considered necessary to slow the spread of the SARS-CoV-2 virus and to mitigate the effects of COVID-2019, the disease it causes.

Measuring the effectiveness of these measures is important for multiple reasons. During the emergency, this information helps improve and quickly adapt policies and -more relevant in a political sense- is helpful to convince poplulations that it is necessary to adhere to them and that the policies are effective. Finally, over a longer timespan, these data are of scientific interest for studies in epidemology.

This evaluation rests on three types of data, all of which is actively collected at the moment:

  • The first pillar is data on the number of confirmed cases and on the number of performed tests, patients with severe disease progression and number of recovered patients. This is part of another challenge of this hackathon. This data can be considered the dependent variable and can be used directly (with the usual caveats regarding specificity, sensitivity, sampling bias etc.) to measure the effect of the measures.
  • The second pillar is auxiliary information such as data on mobility, air quality, mobile phone location data (eg https://github.com/statistikZH/monitoring_covid19/blob/master/Mobility_GooglePopularTimeGraph.md) and data on consumer behavior. This information doesn't directly correlate with case data, however it can be used to infer the adhesion of the general population to the imposed measures regardless of their disease status.
  • The third pillar -and the main effort spent in this project so far- is the collection of a reliable, machine readable data set on all the measures taken by authorities and organisations to slow the spread of SARS-CoV-2. This data is the independent variable -or regressor- to be used in later modeling attempt.

So far the primary outcome of the project has been a crowdsourced collection of all measures taken to slow down the spread of the virus. The data is stored in this repository: https://github.com/baffelli/covid-2019-measures/

The repository is open to contribution, contact Simone Baffelli (https://twitter.com/bafe).

Collecting and encoding the measures taken against the spread of COVID2019

Introduction

Here we collect a timeline of all containment and public health measure taken in Switzerland in the course of the COVID2019 pandemics. This is not of immediate use but we believe it will be very percious for future studies on the effect of different control strategies. We know this is boring work, but we are grateful to all contributors even if they only contributed to one or two rows of data.

Motivation

The federal council has announced the temporary closure of schools and public places non essential to daily life, ordered the cancellation of mass events and banned the gathering of more than 5 persons. These measures follow -and partially replicate at a federal level- similar measures taken earlier by local authorities and other organisations. These strict measures are considered necessary to slow the spread of the SARS-CoV-2 virus and to mitigate the effects of COVID-2019, the disease it causes.

Measuring the effectiveness of these measures is important for multiple reasons. During the emergency, this information helps improve and quickly adapt policies and -more relevant in a political sense- is helpful to convince poplulations that it is necessary to adhere to them and that the policies are effective. Finally, over a longer timespan, these data are of scientific interest for studies in epidemology.

This evaluation rests on three types of data, all of which is actively collected at the moment:

  • The first pillar is data on the number of confirmed cases and on the number of performed tests, patients with severe disease progression and number of recovered patients. This is part of another challenge of this hackathon. This data can be considered the dependent variable and can be used directly (with the usual caveats regarding specificity, sensitivity, sampling bias etc.) to measure the effect of the measures.
  • The second pillar is auxiliary information such as data on mobility, air quality, mobile phone location data (eg https://github.com/statistikZH/monitoring_covid19/blob/master/Mobility_GooglePopularTimeGraph.md) and data on consumer behavior. This information doesn't directly correlate with case data, however it can be used to infer the adhesion of the general population to the imposed measures regardless of their disease status.
  • The third pillar -and the main effort spent in this project so far- is the collection of a reliable, machine readable data set on all the measures taken by authorities and organisations to slow the spread of SARS-CoV-2. This data is the independent variable -or regressor- to be used in later modeling attempt.

So far the primary outcome of the project has been a crowdsourced collection of all measures taken to slow down the spread of the virus. The data is stored in this repository.

The repository is open to contribution, contact Simone Baffelli (https://twitter.com/bafe) or Fabienne Krauer ().

Data format

To collect the measures, we settled a simple (long) csv format:

date_implemented date_lifted measure unit level source measure_cat
date when the measure is implemented date when the measure is lifted textual description of the measure identifier of the administrative unit level of the administrative unit (canton, city, federal) url of the source of this measure category of the measure

The data must be stored in long and tidy format: if multiple measures for the same entities are introduced at the same data, repeated entries must be created.

unit should take one of the following values: - federal for federal level measures - canton for cantonal measures - city for measures taken by municipalities - private for measures taken by businesses, events and other organisations

  • unit: administrative unit. For cantons, use ISO abbreviations (2-letter). For cities, use full name in German, French or Italian. For measures taken by private enterprises or organisations, use their full name.

  • level: should take one of the following values: - federal for federal level measures - canton for cantonal measures - city for measures taken by municipalities - private for measures taken by businesses, events and other organisations

  • source: paste the URL of the source so that it can be verified

  • measure_cat: category of the measure according to the swiss epidemics law. If unsure, leave this field empty.

Measure categorization

To fill the measure_cat column, we propose the following categorization:

example category law article
Kontake von infiszierten Personen werden auffindig gemacht und benachrichtigt contact tracing Swiss Epidemics Act Art. 15
Personen mit Kontakt zu Erkrankten oder Verdachsfälle müssen in Quarantäne zu Hause oder woanders quarantine and self-quarantine Swiss Epidemics Act Art. 35a
Erkrankte Personen müssen zu Hause bleiben oder im Spital isoliert werden isolation and self-isolation Swiss Epidemics Act Art. 35b
Veranstaltungen mit mehr als 1000 Personen werden untersagt oder sind bewilligungspflichtig restriction of events Swiss Epidemics Act Art. 40, 2a
Jegliche Veranstaltungen werden untersagt ban of events Swiss Epidemics Act Art. 40, 2a
Schulschliessung school closure Swiss Epidemics Act Art. 40, 2b
Verkürzte Öffnungszeiten von Ämtern, maximal Kapazitäten in Läden werden festelegt restriction of selected private and public institutions Swiss Epidemics Act Art. 40, 2b
Betriebe wie Bars, Restaurants, Kino, Museen werden geschlossen, Ämter werden geschlossen closure of selected private and public institutions Swiss Epidemics Act Art. 40, 2b
Betreten der Seepromenade ist untersagt, wird gesperrt ban of access to private or public buildings or areas Swiss Epidemics Act Art. 40, 2c
Empfehlung zum häufigen Händewaschen, in Ellbogen husten recommendations: hygiene WHO
Bevölkerung wird aufgefordert Abstand zu halten, Massenansammlungen jeder Art zu meiden (beim Einkaufen, ÖV usw.) recommendations: physical distancing WHO
Besuche im Altersheim, Pflegeheim, Behindertenheim, Spital werden verkürzt restriction of access to health and long-term care facilites Swiss Epidemics Act Art. 40, 2b, Art. 19, 2d
Besuche im Altersheim, Pflegeheim, Behindertenheim, Spital sind untersagt ban of access to health and long-term care facilites Swiss Epidemics Act Art. 40, 2b, Art. 19, 2d
Einreise aus dem Ausland nur für Personen mit Arbeitsstelle in der Schweiz restriction of international mobility Swiss Epidemics Act Art. 41
Einreise aus dem Ausland komplett untersagt mit Ausnahme von Schweizer Bürgern ban of international mobility Swiss Epidemics Act Art. 41
Verstorbene müssen speziell behandelt werden (transport, vorbereitung) management of dead bodies Swiss Epidemics Act Art. 46
Erkrankte oder Verdachtsfälle werden getestet, Einrichtung von Testzentren testing Swiss Epidemics Act Art. 36
Operationen und Eingriffe, welche keine Notfälle sind, werden verschoben suspension of elective surgery ?
Wahlen, Abstimmungen, Parlament oder Ratsmeetings werden verschoben suspension of political events ?

TODO

General

  • [ ] add measures for all cantons and federal and city level measures.
  • [ ] code them according to general categories (Veranstaltungsverbot, ...) according to Art. 40 of the Epiemiegesetz
  • [ ] prepare an interactive visualization of these measures

Missing Entities

  • [x] BE
  • [ ] TG
  • [ ] VS
  • [X] SH
  • [x] AG
  • [x] LU
  • [ ] OW
  • [ ] NW
  • [x] UR
  • [ ] AR
  • [ ] AI
  • [ ] GE
  • [ ] NE
  • [x] SO
  • [x] JU
  • [ ] Cities
  • [ ] Private measures (Events, companies, ...)
  • [ ] Federal measures after 20.03.2020

How to contribute

  1. Select an entity where data is still missing or incomplete.
  2. Find the correspoding sheet on the collaborative Google Sheet here. If the sheet does not exist, create it.
  3. For cantonal level, find the corresponding home page of the cantonal administration where recommendations and orders are published. These are usually published as PDF's under "media announcements / Medienmitteilungen"
  4. Extract the information given in these published orders into the Google sheet. Use the language you are most comfortable with Extract the following information:
  5. date_implemented: Date of implementation of the measures. Usually this is the date of the publication. If the publication just announces a future measure, do not register it yet as it is not implemented yet. Format is dd.mm.yyyy
  6. datelifted: Date when the control measures are lifted or undone. In most cases, you will not know this, so leave the field empty. For actions with a clear time horizon (i.e. School closure), the datelifted can be added. However, we will have to check this continuously, since these measures are subject to change. Format is dd.mm.yyyy
  7. measure: Free text, short description of measures taken. Look for any recommendation or action/orders issued in relation to stopping the transmission. Examples are given below.
  8. measure_cat: category of measure. These are coded according to measures described in the Federal Pandemic law and according to the recommendations of WHO. If you are unsure about the category, leave the field empty. Use only the categories provided, do not make new categories.

Updated 21:48 25.03.2020
Maintained by bafe

  • 11:10 25.03.2020 / oleg / update
  • 11:09 25.03.2020 / oleg / update
  • 09:29 24.03.2020 / bafe / update
  • 09:29 24.03.2020 / bafe / update
  • 09:17 24.03.2020 / bafe / update

Connect to the community on Chat | Forum | Twitter

All attendees, sponsors, partners, volunteers and staff at our hackathon are required to agree with the Hack Code of Conduct. Organisers will enforce this code throughout the event. We expect cooperation from all participants to ensure a safe environment for everybody. For more details on how the event is run, see the Remote Guidelines.

Creative Commons LicenceThe contents of this website, unless otherwise stated, are licensed under a Creative Commons Attribution 4.0 International License.