docs: Major update of risk assessment document, incl. example

This commit is contained in:
Sebastian Wolf 2020-07-09 18:40:28 +02:00
parent 2f16aa4e4f
commit 0efba5e1a0
No known key found for this signature in database
GPG Key ID: B53C9059C9EEFDC8
3 changed files with 267 additions and 30 deletions

View File

@ -13,3 +13,7 @@ solution_architecture*.md @tklingbeil @SebastianWolf-SAP @tkowark @LukasMasuch @
# Epidemiological Motivation of the Transmission Risk Level
transmission_risk* @kirchnergo @hoehleatsu @weidemannf @dirkschumacher @benzlerj
# How does the Corona-Warn-App identify an increased risk?
cwa-risk* @benzlerj @hoehleatsu @kirchnergo @SebastianWolf-SAP
translations/cwa-risk* @benzlerj @hoehleatsu @kirchnergo @SebastianWolf-SAP

View File

@ -2,41 +2,157 @@
## Prerequisites
People who use the Corona-Warn-App (CWA) and are tested positive for the SARS-CoV-2 coronavirus can allow their CWA to upload the random device keys (*temporary exposure keys*) that have been generated on their smartphones in recent days to the Corona-Warn-App server as *diagnosis keys* and release them there. These diagnosis keys are the basis for risk identification for all other CWA users.
People who use the Corona-Warn-App (CWA) and are tested positive for the SARS-CoV-2 coronavirus can allow their CWA to upload the random device keys (*temporary exposure keys*) that have been generated by their smartphone's operating system in recent days to the Corona-Warn-App server as *diagnosis keys* and release them there.
These diagnosis keys are the basis for risk identification for all other CWA users.
A user who has tested positive for coronavirus uploads up to 15 diagnosis keys: one for each of the up to 14 days before the download, as well as one for the current day, which is uploaded the next day. The latter is necessary because diagnosis keys can only be uploaded for past days.
A user who has tested positive for coronavirus uploads up to 15 diagnosis keys: one for each of the up to 14 days before the upload, as well as (still to be implemented) one for the current day, which is uploaded the next day.
The latter is necessary because diagnosis keys can only be uploaded for past days in order to prevent abuse of diagnosis keys that are still active.
Diagnosis keys do not give any indication as to the identity of a person who has tested positive, but a diagnosis key from a certain day can be matched with the *rolling proximity identifiers* that a users smartphone has transmitted via Bluetooth throughout a given day and were received and recorded by other smartphones nearby. Each diagnosis key is appended with a value that indicates the *transmission risk level* that likely existed for the person who has tested positive on the day that the diagnosis key belongs to. This transmission risk is then estimated in a complex procedure, which is based on empirical values and takes the latest scientific findings into account. Every diagnosis key expires after 14 days. Therefore, only the diagnosis keys from the last 14 days are available.
Diagnosis keys do not give any indication as to the identity of a person who has tested positive, but a diagnosis key from a certain day can be matched with the *rolling proximity identifiers* that a users smartphone has transmitted via Bluetooth throughout a given day and were received and recorded by other smartphones nearby.
Each diagnosis key is appended with a value that indicates the *transmission risk level* that likely existed for the person who has tested positive on the day that the diagnosis key belongs to.
This transmission risk is [estimated in a mathematical procedure](transmission_risk.pdf), which is based on empirical evidence and takes the latest scientific findings into account.
Every diagnosis key expires after 14 days. Therefore, only the diagnosis keys from the last 14 days are available.
## Procedure
Several times per day, all active Corona-Warn-Apps download the diagnosis keys released on the Corona-Warn-App server and pass them on to the operating system in batches through an interface. The app checks whether any of these received, recorded rolling proximity identifiers match any of the diagnosis keys. If there is a match, this indicates that the users smartphone encountered the smartphone of a person who has uploaded a diagnosis key on the day to which the diagnosis key belongs.
Daily, all active Corona-Warn-Apps download the diagnosis keys released on the Corona-Warn-App server and pass them on to the smartphone's operating system in batches through an interface.
The operating system checks whether any of the rolling proximity identifiers that it has received and recorded over the past 14 days match any of the diagnosis keys.
If there is a match, this indicates that the users smartphone encountered the smartphone of a person who has uploaded the diagnosis key, on the day to which the diagnosis key belongs.
In the next step, the app analyzes all the matching rolling proximity identifiers for each diagnosis key, to estimate how long the exposure lasted in total on the day in question and how close the smartphones were to each other on average during the exposure. The distance is calculated from the measured reduction in strength of the Bluetooth signal, which is specified in dB (decibel). All exposures for a diagnosis key that lasted less than 10 minutes in total (regardless of how close the smartphones came during that time) or during which the smartphones were more than 8 meters (73 dB attenuation) apart on average (regardless of how long the exposure lasted) are discarded as harmless.
> Note: Days in the context of the Corona-Warn-App and thus in this document are calendar days according to UTC (Coordinated Universal Time). They change at 1 am Central European Time and 2 am Central European Summer Time, respectively.
> NB: In the following, the total of all exposures that belong to a diagnosis key, that is, all exposures over a day between the same two smartphones, is referred to as the “exposure set”.
In the next step, the operating system analyzes all the matching rolling proximity identifiers for each diagnosis key, to estimate how long the encounter lasted in total on the day in question and how close both smartphones were to each other on average during the encounter.
The distance is calculated from the measured reduction in strength of the Bluetooth signal, which is specified in dB (decibel).
Under perfect circumstances, i.e. without any obstacle in the signal pathway (see also section “Consequences and Constraints”), each dB value is associated with a particular distance.
All encounters for a diagnosis key that lasted less than 10 minutes in total (regardless of how close the smartphones came during that time) or during which the smartphones were more than 8 meters (>73 dB attenuation) apart on average (regardless of how long the encounter lasted) are discarded as negligible risk.
For the remaining exposures that have not been discarded as harmless, a *total risk score* is calculated for each exposure set, by multiplying the transmission risk score described above by the *days since last exposure value*, which is calculated as the time between the day of the last exposure and the current day.
> Note: In the following, the total of all encounters that belong to a particular diagnosis key, that is, all encounters over a given day between the same two smartphones, is referred to as the “encounter set”.
All exposure sets that exceed a certain threshold (the *minimum risk score*) are considered to be risk exposures. The other exposure sets are discarded as harmless, like the sets that were previously discarded for being too short and/or too distant.
For the remaining encounters that have not been discarded, a *total risk score* is calculated for each encounter set, by multiplying the transmission risk score described above by the *days since last exposure value*, which is derived from the day count from the last encounter to the current day.
At the same time, the remaining risk exposures are added together to determine how much time exposure took place within a very close range below 1.5 meters (55 dB attenuation) and how much time exposure took place in a close range between 1.5 and 3 meters (63 dB attenuation).
All encounter sets whose total risk score exceeds a certain threshold (the *minimum risk score*) are considered to be risk exposures.
The other encounter sets are discarded as negligible risk, like the sets that were previously discarded for being too short and/or too distant.
The total calculated time is then cross- calculated against the *maximum risk score*, the exposure with the highest risk: the time remains unchanged if this risk is estimated as average (for risk exposures), it is extended to one and a half times if the risk is above average, and it is reduced significantly (to around one-sixth) if the risk is below average. As a result, an exposure time of 10 minutes can be extended to more than 15 minutes and an exposure time of 45 minutes can be reduced to less than 10 minutes.
At the same time, all remaining risk exposures are added together to determine how much time exposure took place within a very close range below 1.5 meters (<55 dB attenuation) and how much time exposure took place in a close range between 1.5 and 3 meters (55 to 63 dB attenuation).
Time spent in a distance greater than 3 meters apart will not be considered.
The total calculated time of all exposures of the last 14 days is then adjusted using the *maximum risk score* of the exposure with the highest risk: the time remains unchanged if this risk is estimated as average (for risk exposures), it is extended to one and a half times if the risk is above average, and it is reduced significantly (to around one-sixth) if the risk is below average.
As a result, an exposure time of 10 minutes can be extended to more than 15 minutes and an exposure time of 45 minutes can be reduced to less than 10 minutes.
## Consequences and Constraints
In the end, a CWA user is notified of an increased risk whenever the risk exposure time calculated as described above amounts to 15 minutes or longer. This notification takes place in the CWA and, at the same time, provides recommendations as to how the user should proceed.
In the end, a CWA user is notified of an increased risk whenever the risk exposure time calculated as described above amounts to 15 minutes or longer.
This notification takes place in the CWA and, at the same time, provides recommendations as to how the user should proceed.
When assessing the times and distances calculated by the CWA, it is important to consider that it is not possible to measure these two parameters precisely. The individually measured times can deviate from the actual exposure time by 5 minutes plus or minus and the calculated distances are approximate values under ideal conditions, that is, without any impediments between the two smartphones. Even minor impediments, such as a person between the two smartphones or a signal-impeding smartphone case, can cause the distance to appear to be twice as large as it actually is.
When assessing the times and distances calculated by the CWA, it is important to consider that it is not possible to measure these two parameters precisely.
The individually measured times can deviate from the actual encounter time by 5 minutes plus or minus and the calculated distances are approximate values under ideal conditions, that is, without any impediments between the two smartphones.
Even minor impediments, such as a person between the two smartphones or a signal-impeding smartphone case, can cause the distance to appear to be twice as large as it actually is.
Due to privacy considerations, the properties described above can currently only be queried for the total set of all risk exposures at the interface to the operating system, but not for individual risk exposures or exposure by day. As long as the number of new infections remains relatively low, this should not make much of a difference, because it is likely that only very few CWA users will have been exposed to multiple persons who have tested positive within the time frame until they are notified.
Due to privacy considerations, the properties described above can currently only be queried for the total set of all risk exposures at the interface to the operating system, but not for individual risk exposures or exposure by day.
As long as the number of new infections remains relatively low, this should not make much of a difference, because it is likely that only very few CWA users will have been exposed to multiple persons who have tested positive within the time frame until they are notified.
## Current configuration
## An Example
As documented in the [risk score calculation section](https://github.com/corona-warn-app/cwa-documentation/blob/master/solution_architecture.md#risk-score-calculation) of the solution architecture document, the actual parameters for the calculation are provided by a set of parameters which are hosted on cwa-server. This configuration might change over time according to the latest research results. The respective current set of configuration values can be looked up in the [cwa-server repository](https://github.com/corona-warn-app/cwa-server):
Anton and Aisha are each notified on the 20th of the month that according to their test results, they have contracted COVID-19.
That same day, Anton allows his CWA to notify other CWA users with whom he has had risk exposures.
The CWA has been running on his smartphone continuously for the past week.
The CWA now uploads his temporary exposure keys from the last 7 days to the CWA server as diagnosis keys (no further keys are available, because Anton has only been using the CWA for 8 days and the current exposure keys cannot be uploaded yet).
They are assigned the transmission risk levels VI (for the previous day), three times VIII (for the 16th to the 18th), V, III, and I (in descending order for the other past days, the 13th to the 15th).
|||||||||
|-|:-:|:-:|:-:|:-:|:-:|:-:|:-:|
|Transmission risk level|VI|VIII|VIII|VIII|V|III|I|
|Days until upload consent granted|1|2|3|4|5|6|7|
|Generation date of the key|19th|18th|17th|16th|15th|14th|13th|
Table 1: Transmission risk level for Antons 7 shared diagnosis keys, based on the interval from the day upload consent is granted (20th)
Aisha hesitates for a day and does not grant consent until the 21st of the month.
Since she activated her CWA several weeks ago and has been running it in the background ever since, her temporary exposure keys from the last 14 days are available to upload.
Her diagnosis keys are also assigned the transmission risk level VI, three times VIII, V, III, and I in descending order for the previous seven days, but starting with the 20th, and thus offset one day compared to Anton.
(The CWA does not know that both people were informed of their test results on the same day. In the current version, only the date on which consent to upload is granted is available to determine the day-specific transmission risk levels.)
The 7 older days, the period from the 7th to the 13th of the month, are each assigned the transmission risk level I.
||||||||||||||||
|-|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|
|Transmission risk level|VI|VIII|VIII|VIII|V|III|I|I|I|I|I|I|I|I|
|Days until upload consent granted|1|2|3|4|5|6|7|8|9|10|11|12|13|14|
|Generation date of the key|20th|19th|18th|17th|16th|15th|14th|13th|12th|11th|10th|9th|8th|7th|
Table 2: Transmission risk level for Aishas 14 shared diagnosis keys, based on the interval from the day upload consent is granted (21st)
Anton and Aisha regularly travel to work together.
Betty takes the same route and occasionally rides the same bus.
All three of them use their smartphones during the journey, which means there are no impediments to the Bluetooth signals.
Betty met Anton and Aisha on the 9th and the 16th, for 10 minutes each in the morning and in the evening.
Anton sat one meter away from Betty, while Aisha sat and two meters away.
When Bettys CWA retrieves Antons diagnosis keys on the 21st and passes them on to her smartphone's operating system's interface, an encounter set is recognized for the 16th.
(Antons CWA did not upload a diagnosis key for the 9th.)
This encounter set lasted a total of 20 minutes and the smartphones were an average of one meter apart.
This results in values of 1 for both duration and attenuation (also see “Exposure Configuration” in the section “Current Configuration”).
This ensures that this encounter set is not discarded.
The value of the parameter for the delay risk (21 - 16 = 5 days) is configured at 5 continuously and the value for the transmission risk is taken directly from the transmission risk level, which means it is 8.
The total risk level is therefore calculated as 1 x 1 x 5 x 8 = 40, which incidentally is the highest value that can be reached with the current configuration.
The threshold of 11 is exceeded, which means the encounter set counts as a risk exposure.
| | | | | | | | | |
|-|-|-|-|-|-|-|-|-|
|Days since last exposure| >=14d (5) | 12-13d (5) | 10-11d (5) | 8-9d (5) | 6-7d (5) | **4-5d (5)** | 2-3d (5) | 0-1d (5) |
|Attenuation| >73dB (0) | >63-<=73dB (1) | >51-<=63dB (1) | **>33-<=51dB (1)** | >27-<=33dB (1) | >15-<=27dB (1) | >10-<=15dB (1) | <=10dB (1) |
|Duration| 0min (0) | >0-<=5min (0) | >5-<=10min (0) | >10-<=15min (1) | **>15-<=20min (1)** | >20-<=25min (1) | >25-<=30min (1) | >30min (1) |
|Transmission risk| I (1) | II (2) | III (3) | IV (4) | V (5) | VI (6) | VII (7) | **VIII (8)** |
Table 3: Parameter values for Bettys encounter set with Anton on the 16th
Since this risk exposure is the only risk exposure known to Bettys CWA, it is the only one taken into account in the summary evaluation of her times spent in the distance ranges up to 1.5 meters and up to 3 meters.
Betty spent 20 minutes in the distance range below 1.5 meters, which are counted fully.
Even if these 20 minutes are cross-calculated against the risk exposure with the highest risk, only this one risk exposure is taken into account, with its exposure risk of 40.
The multiplication of these 20 minutes by 40/25 (25 is the currently configured value for “average risk” exposures; also see the risk score normalization divisor in "Configuration of Attenuation and Duration” in the section “Current Configuration”) equals 32 minutes.
Since the CWA sends notifications of increased risk starting with 15 minutes, Bettys app sends her a notification.
She is notified that she had a risk exposure 5 days ago.
The next day, the 22nd, Bettys CWA also retrieves Aishas diagnosis keys.
It identifies additional encounters on the 16th and the 9th of the month.
Both encounter sets lasted a total of 20 minutes each and the smartphones were an average of two meters apart.
This also results in values of 1 for duration and attenuation.
The delay risk values (22 - 16 = 6 days; 22 - 9 = 13 days) are a constant 5 and the transmission risk values are 5 for the 16th and 1 for the 9th of the month.
As a result, the exposure risks for the 16th are calculated as 1 x 1 x 5 x 5 = 25 and for the 9th as 1 x 1 x 5 x 1 = 5.
The encounter set from the 9th does not reach the threshold and is therefore not counted as a risk exposure.
| | | | | | | | | |
|-|-|-|-|-|-|-|-|-|
|Days since last exposure| >=14d (5) | 12-13d (5) | 10-11d (5) | 8-9d (5) | **6-7d (5)** | 4-5d (5) | 2-3d (5) | 0-1d (5) |
|Attenuation| >73dB (0) | >63-<=73dB (1) | **>51-<=63dB (1)** | >33-<=51dB (1) | >27-<=33dB (1) | >15-<=27dB (1) | >10-<=15dB (1) | <=10dB (1) |
|Duration| 0min (0) | >0-<=5min (0) | >5-<=10min (0) | >10-<=15min (1) | **>15-<=20min (1)** | >20-<=25min (1) | >25-<=30min (1) | >30min (1) |
|Transmission risk| I (1) | II (2) | III (3) | IV (4) | **V (5)** | VI (6) | VII (7) | VIII (8) |
Table 4: Parameter values for Bettys encounter set with Aisha on the 16th
| | | | | | | | | |
|-|-|-|-|-|-|-|-|-|
|Days since last exposure| >=14d (5) | **12-13d (5)** | 10-11d (5) | 8-9d (5) | 6-7d (5) | 4-5d (5) | 2-3d (5) | 0-1d (5) |
|Attenuation| >73dB (0) | >63-<=73dB (1) | **>51-<=63dB (1)** | >33-<=51dB (1) | >27-<=33dB (1) | >15-<=27dB (1) | >10-<=15dB (1) | <=10dB (1) |
|Duration| 0min (0) | >0-<=5min (0) | >5-<=10min (0) | >10-<=15min (1) | **>15-<=20min (1)** | >20-<=25min (1) | >25-<=30min (1) | >30min (1) |
|Transmission risk| **I (1)** | II (2) | III (3) | IV (4) | V (5) | VI (6) | VII (7) | VIII (8) |
Table 5: Parameter values for Bettys encounter set with Aisha on the 9th
In contrast, the risk exposure on the 16th is taken into account in the summary evaluation, which means Bettys CWA now counts 20 minutes (with Anton) in the distance range up to 1.5 meters fully and an additional 20 minutes (with Aisha) in the distance range up to 3 meters half (that is, 10 minutes).
The calculated 30 minutes are once again cross-calculated with Bettys risk exposure with Anton, which at 40 is still the highest exposure risk from all risk encounters identified in the recorded 14-day period, resulting in 30 x 40/25 = 48 minutes.
Bettys updated risk notification now shows 2 risk encounters, the last of which took place 6 days ago.
## Current Configuration
As documented in the [risk score calculation section](https://github.com/corona-warn-app/cwa-documentation/blob/master/solution_architecture.md#risk-score-calculation) of the solution architecture document, the actual parameters for the calculation are provided by a set of parameters which are hosted on cwa-server.
This configuration might change over time according to the new research results and insights. The respective current set of configuration values can be looked up in the [cwa-server repository](https://github.com/corona-warn-app/cwa-server):
- [Exposure Configuration](https://github.com/corona-warn-app/cwa-server/blob/master/services/distribution/src/main/resources/master-config/exposure-config.yaml)
- [Attenuation & Duration Configuration](https://github.com/corona-warn-app/cwa-server/blob/master/services/distribution/src/main/resources/master-config/attenuation-duration.yaml)
- [App Configuration, including minimum risk score](https://github.com/corona-warn-app/cwa-server/blob/master/services/distribution/src/main/resources/master-config/app-config.yaml)
- [Risk Score Classification](https://github.com/corona-warn-app/cwa-server/blob/master/services/distribution/src/main/resources/master-config/risk-score-classification.yaml)
- [Risk Score Classification](https://github.com/corona-warn-app/cwa-server/blob/master/services/distribution/src/main/resources/master-config/risk-score-classification.yaml)

View File

@ -2,41 +2,158 @@
## Voraussetzungen
Personen, die die Corona-Warn-App (CWA) nutzen und positiv auf das Coronavirus SARS-CoV-2 getestet wurden, können ihrer CWA erlauben, die vom Betriebssystem ihres Smartphones erzeugten zufälligen Geräteschlüssel (*Temporary Exposure Keys*) der vergangenen Tage als sogenannte Positivkennungen (*Diagnosis Keys*) auf den Corona-Warn-App-Server hochzuladen und dort zu veröffentlichen. Diese Positivkennungen sind die Grundlage der Risikoermittlung für alle anderen die CWA nutzenden Personen.
Personen, die die Corona-Warn-App (CWA) nutzen und positiv auf das Coronavirus SARS-CoV-2 getestet wurden, können ihrer CWA erlauben, die vom Betriebssystem ihres Smartphones erzeugten zufälligen Geräteschlüssel (*Temporary Exposure Keys*) der vergangenen Tage als sogenannte Positivkennungen (*Diagnosis Keys*) auf den Corona-Warn-App-Server hochzuladen und dort zu veröffentlichen.
Diese Positivkennungen sind die Grundlage der Risikoermittlung für alle anderen die CWA nutzenden Personen.
Eine Corona-positiv getestete Person lädt bis zu 15 Positivkennungen hoch, nämlich je eine für jeden der bis zu 14 letzten Tage vor dem Hochladen sowie am folgenden Tag die für den aktuellen Tag. Letzteres ist nötig, da Positivkennungen nur für bereits vergangene Tage hochgeladen werden können.
Eine Corona-positiv getestete Person lädt bis zu 15 Positivkennungen hoch, nämlich je eine für jeden der bis zu 14 letzten Tage vor dem Hochladen sowie (derzeit noch nicht umgesetzt) am folgenden Tag die für den aktuellen Tag.
Letzteres ist nötig, da Positivkennungen nur für bereits vergangene Tage hochgeladen werden können, um Missbrauch noch aktiver Positivkennungen zu verhindern.
Positivkennungen lassen keine Rückschlüsse auf die Identität der positiv getesteten Person zu, aber die Positivkennung eines bestimmten Tages passt zu den ständig wechselnden Zufallscodes (*Rolling Proximity Identifiers*), die das Smartphone der Person den ganzen Tag über mittels Bluetooth versendet hat und die von den Smartphones in der Nähe empfangen und aufgezeichnet wurden. An jede Positivkennung ist noch ein Wert angehängt, der angibt, wie groß das Übertragungsrisiko (*Transmission Risk Level*) der positiv getesteten Person an dem Tag, zu dem die Positivkennung gehört, vermutlich gewesen ist. Dieses Übertragungsrisiko wird anhand von Erfahrungswerten unter Berücksichtigung der aktuell vorliegenden wissenschaftlichen Erkenntnisse in einem komplizierten Verfahren geschätzt. Jede Positivkennung verfällt, wenn sie älter als 14 Tage ist. Deshalb sind stets nur die Positivkennungen der letzten 14 Tage verfügbar.
Positivkennungen lassen keine Rückschlüsse auf die Identität der positiv getesteten Person zu, aber die Positivkennung eines bestimmten Tages passt zu den ständig wechselnden Zufallscodes (*Rolling Proximity Identifiers*), die das Smartphone der Person den ganzen Tag über mittels Bluetooth versendet hat und die von anderen Smartphones in der Nähe empfangen und aufgezeichnet wurden.
An jede Positivkennung ist noch ein Wert angehängt, der angibt, wie groß das Übertragungsrisiko (*Transmission Risk Level*) der positiv getesteten Person an dem Tag, zu dem die Positivkennung gehört, vermutlich gewesen ist. Dieses Übertragungsrisiko wird anhand von Erfahrungswerten unter Berücksichtigung der aktuell vorliegenden wissenschaftlichen Erkenntnisse [in einem mathematischen Verfahren geschätzt](../transmission_risk.pdf).
Jede Positivkennung verfällt, wenn sie älter als 14 Tage ist. Deshalb sind stets nur die Positivkennungen der letzten 14 Tage verfügbar.
## Verfahren
Alle aktiven Corona-Warn-Apps laden täglich vom Corona-Warn-App-Server die dort veröffentlichten Positivkennungen herunter und übergeben sie gesammelt über eine Schnittstelle an das Betriebssystem. Dort wird geprüft, ob empfangene und aufgezeichnete Zufallscodes vorliegen, die zu einer der Positivkennungen passen. Ein solches Zusammenpassen zeigt an, dass sich das Smartphone, das die Zufallscodes aufgezeichnet hat, und das Smartphone der Corona-positiv getesteten Person, die die Positivkennung hochgeladen hat, an dem Tag, zu dem die Positivkennung gehört, begegnet sind.
Alle aktiven Corona-Warn-Apps laden täglich vom Corona-Warn-App-Server die dort veröffentlichten Positivkennungen herunter und übergeben sie gesammelt über eine Schnittstelle an das Betriebssystem des Smartphones.
Dort wird geprüft, ob empfangene und aufgezeichnete Zufallscodes der letzten 14 Tage vorliegen, die zu einer der Positivkennungen passen.
Ein solches Zusammenpassen zeigt an, dass sich das Smartphone, das die Zufallscodes aufgezeichnet hat, und das Smartphone der Corona-positiv getesteten Person, die die Positivkennung hochgeladen hat, an dem Tag, zu dem die Positivkennung gehört, begegnet sind.
Als nächstes wird für jede Positivkennung anhand aller dazu passenden Zufallscodes geschätzt, wie lange die Begegnungen jenes Tags insgesamt gedauert haben und wie nahe die Smartphones sich dabei im Durchschnitt waren. Die Entfernung wird aus der gemessenen Abschwächung des Bluetooth-Signals errechnet, die in dB (Dezibel) angegeben wird. Jedem dB-Wert lässt sich eine Entfernung im Freiraum (d.h., ohne Hindernisse im Signalweg; s.a. Erläuterungen im Abschnitt "Folgen und Einschränkungen") zuordnen. Alle Begegnungen zu einer Positivkennung, die insgesamt weniger als 10 Minuten gedauert haben (egal, wie nahe sich die Smartphones dabei gekommen sind) oder bei denen die Smartphones im Durchschnitt mehr als ca. 8 Meter Freiraum (73 dB Dämpfung) voneinander entfernt waren (egal, wie lange sie insgesamt gedauert haben), werden als unbedenklich verworfen.
> NB: Tage im Kontext der Corona-Warn-App und damit auch dieses Dokuments sind Kalendertage nach UTC (Coordinated Universal Time). Der Tageswechsel erfolgt demnach um 1 Uhr Mitteleuropäischer Zeit bzw. 2 Uhr Mitteleuropäischer Sommerzeit.
> NB: Wir bezeichnen die Gesamtheit aller Begegnungen, die jeweils zu einer Positivkennung gehören, also alle Begegnungen eines Tages zwischen denselben zwei Smartphones, im Weiteren als Begegnungsmenge.
Als nächstes wird im Betriebssystem des Smartphones für jede Positivkennung anhand aller dazu passenden Zufallscodes geschätzt, wie lange die Begegnungen jenes Tags insgesamt gedauert haben und wie nahe die beiden Smartphones sich dabei im Durchschnitt waren.
Die Entfernung wird aus der gemessenen Abschwächung des Bluetooth-Signals errechnet, die in dB (Dezibel) angegeben wird.
Jedem dB-Wert lässt sich eine Entfernung im Freiraum (d.h., ohne Hindernisse im Signalweg; s.a. Erläuterungen im Abschnitt "Folgen und Einschränkungen") zuordnen.
Alle Begegnungen zu einer Positivkennung, die insgesamt weniger als 10 Minuten gedauert haben (egal, wie nahe sich die Smartphones dabei gekommen sind) oder bei denen die Smartphones im Durchschnitt mehr als ca. 8 Meter Freiraum (>73 dB Dämpfung) voneinander entfernt waren (egal, wie lange sie insgesamt gedauert haben), werden als unbedenklich verworfen.
> NB: Wir bezeichnen die Gesamtheit aller Begegnungen, die jeweils zu einer bestimmten Positivkennung gehören, also alle Begegnungen eines Tages zwischen denselben zwei Smartphones, im Weiteren als Begegnungsmenge.
Bei den restlichen, nicht als unbedenklich verworfen Begegnungen wird für jede Begegnungsmenge ein Begegnungsrisiko (*Total Risk Score*) errechnet, indem der oben erläuterte Übertragungsrisikowert mit einem Verzugsrisikowert (*Days Since Last Exposure Value*) multipliziert wird, der sich aus der Anzahl der seit der Begegnung vergangenen Tage ableitet.
Alle Begegnungsmengen, die dabei einen bestimmten Grenzwert (*Minimum Risk Score*) überschreiten, werden als Risikobegegnungen angesehen. Die anderen Begegnungsmengen werden ebenso wie zuvor schon die zu kurzen oder zu entfernten Begegnungsmengen als unbedenklich verworfen.
Alle Begegnungsmengen, deren Begegnungsrisiko dabei einen bestimmten Grenzwert (*Minimum Risk Score*) überschreitet, werden als Risikobegegnungen angesehen.
Die anderen Begegnungsmengen werden ebenso wie zuvor schon die zu kurzen oder zu entfernten Begegnungsmengen als unbedenklich verworfen.
Zugleich wird für alle verbleibenden Begegnungsmengen, die Risikobegegnungen, zusammengezählt, wieviel Zeit in einem sehr nahen Entfernungsbereich unter ca. 1,5 Metern (55 dB Dämpfung) verbracht wurde und wieviel Zeit in einem nahen Entfernungsbereich zwischen ca. 1,5 und 3 Metern (63 dB Dämpfung) verbracht wurde. Dabei wird die Zeit im sehr nahen Bereich ganz und die Zeit im nahen Bereich zur Hälfte gezählt.
Zugleich wird für alle verbleibenden Begegnungsmengen, die Risikobegegnungen, zusammengezählt, wieviel Zeit in einem sehr nahen Entfernungsbereich unter ca. 1,5 Metern (<55 dB Dämpfung) verbracht wurde und wieviel Zeit in einem nahen Entfernungsbereich zwischen ca. 1,5 und 3 Metern (55 bis 63 dB Dämpfung) verbracht wurde.
Dabei wird die Zeit im sehr nahen Bereich ganz und die Zeit im nahen Bereich zur Hälfte gezählt.
Die in einer Entfernung von mehr als ca. 3 Metern verbrachte Zeit wird nicht gezählt.
Die so errechnete Gesamtzeit wird dann noch mit dem Begegnungsrisiko der Risikobegegnung mit dem höchsten Risiko (*Maximum Risk Score*) verrechnet. Und zwar so, dass sie unverändert bleibt, wenn dieses Risiko als durchschnittlich (für Risikobegegnungen) eingeschätzt wird, dass sie sich bis auf das ungefähr anderthalbfache verlängert, wenn dieses Risiko überdurchschnittlich ist, und deutlich (bis auf ungefähr ein Sechstel) verkürzt, wenn dieses Risiko unterdurchschnittlich ist. Dadurch kann eine Zeit, die zuvor 10 Minuten betrug, auf über 15 Minuten verlängert werden und eine Zeit, die zuvor 45 Minuten betrug, auf unter 10 Minuten verkürzt werden.
Die so errechnete Gesamtzeit aller Risikobegegnungen der letzten 14 Tage wird dann noch mit dem Begegnungsrisiko der Risikobegegnung mit dem höchsten Risiko (*Maximum Risk Score*) verrechnet.
Und zwar so, dass sie unverändert bleibt, wenn dieses Risiko als durchschnittlich (für Risikobegegnungen) eingeschätzt wird, dass sie sich bis auf das ungefähr anderthalbfache verlängert, wenn dieses Risiko überdurchschnittlich ist, und deutlich (bis auf ungefähr ein Sechstel) verkürzt, wenn dieses Risiko unterdurchschnittlich ist.
Dadurch kann eine Zeit, die zuvor 10 Minuten betrug, auf über 15 Minuten verlängert werden und eine Zeit, die zuvor 45 Minuten betrug, auf unter 10 Minuten verkürzt werden.
## Folgen und Einschränkungen
Am Ende wird die die CWA nutzende Person immer dann über ein erhöhtes Risiko benachrichtigt, wenn die so ermittelte gesamte Risikobegegnungszeit 15 Minuten oder länger ist. Diese Benachrichtigung erfolgt in der CWA und gibt der Person zugleich Handlungsempfehlungen für das weitere Vorgehen.
Am Ende wird die die CWA nutzende Person immer dann über ein erhöhtes Risiko benachrichtigt, wenn die so ermittelte gesamte Risikobegegnungszeit 15 Minuten oder länger ist.
Diese Benachrichtigung erfolgt in der CWA und gibt der Person zugleich Handlungsempfehlungen für das weitere Vorgehen.
Bei der Bewertung der von der CWA ermittelten Zeiten und Entfernungen muss berücksichtigt werden, dass beide Parameter nicht exakt gemessen werden können. Die einzelnen gemessenen Zeiten können bis zu 5 Minuten in beide Richtungen abweichen und die ermittelten Entfernungen sind Näherungswerte unter Idealbedingungen, d.h., wenn z.B. keine Hindernisse zwischen den beiden Smartphones sind. Schon bei kleineren Hindernissen, z.B. einer Person zwischen den beiden Smartphones oder wenn das Smartphone abgeschirmt verstaut ist, kann die Entfernung doppelt so groß erscheinen wie sie wirklich ist.
Bei der Bewertung der von der CWA ermittelten Zeiten und Entfernungen muss berücksichtigt werden, dass beide Parameter nicht exakt gemessen werden können.
Die einzelnen gemessenen Zeiten können bis zu 5 Minuten in beide Richtungen abweichen und die ermittelten Entfernungen sind Näherungswerte unter Idealbedingungen, d.h., wenn z.B. keine Hindernisse zwischen den beiden Smartphones sind.
Schon bei kleineren Hindernissen, z.B. einer Person zwischen den beiden Smartphones oder wenn das Smartphone abgeschirmt verstaut ist, kann die Entfernung doppelt so groß erscheinen wie sie wirklich ist.
Aufgrund von Datenschutzerwägungen können die oben beschriebenen Eigenschaften an der Schnittstelle zum Betriebssystem vorerst nur für die Gesamtheit aller Risikobegegnungen abgefragt werden, nicht aber für einzelne Risikobegegnungen oder tageweise. Solange die Anzahl neuer Fälle vergleichsweise gering bleibt, dürfte das keinen großen Unterschied machen, da vermutlich nur wenige die CWA nutzende Personen im Zeitraum vor ihrer Benachrichtigung Risikobegegnungen mit mehreren positiv getesteten Personen haben, die die CWA nutzen.
Aufgrund von Datenschutzerwägungen können die oben beschriebenen Eigenschaften an der Schnittstelle zum Betriebssystem vorerst nur für die Gesamtheit aller Risikobegegnungen abgefragt werden, nicht aber für einzelne Risikobegegnungen oder tageweise.
Solange die Anzahl neuer Fälle vergleichsweise gering bleibt, dürfte das keinen großen Unterschied machen, da vermutlich nur wenige die CWA nutzende Personen im Zeitraum vor ihrer Benachrichtigung Risikobegegnungen mit mehreren positiv getesteten Personen haben, die die CWA nutzen.
## Ein Beispiel
Anton und Aisha werden am 20. eines Monats über ihr jeweils Corona-positives Testergebnis informiert.
Anton erlaubt noch am selben Tag seiner CWA, andere die CWA nutzende Personen, mit denen er Risikobegegnungen hatte, darüber zu informieren.
Er hat die CWA auf seinem Smartphone seit einer Woche durchgehend aktiviert.
Die CWA lädt nun seine zufälligen Geräteschlüssel der letzten 7 Tage (mehr sind nicht verfügbar, da Anton die CWA erst 8 Tage im Einsatz hat und der aktuelle Geräteschlüssel noch nicht hochgeladen werden kann) als Positivkennungen auf den CWA-Server hoch.
Sie werden mit den Übertragungsrisikograden VI (für den Vortag), dreimal VIII (für den 16. bis 18.), V, III und I (rückwärts für die anderen vorausgegangenen Tage, 13. bis 15.) versehen.
|||||||||
|-|:-:|:-:|:-:|:-:|:-:|:-:|:-:|
|Übertragungsrisikograd|VI|VIII|VIII|VIII|V|III|I|
|Abstand zum Tag der Upload-Zustimmung|1|2|3|4|5|6|7|
|Erzeugungsdatum des Schlüssels|19.|18.|17.|16.|15.|14.|13.|
Tabelle 1: Übertragungsrisikograd für Antons 7 geteilte Positivkennungen, basierend auf dem Abstand zum Tag der Upload-Zustimmung (20.)
Aisha zögert noch einen Tag und gibt ihre Zustimmung erst am 21. des Monats.
Da sie ihre CWA schon vor mehreren Wochen aktiviert hatte und seitdem durchgehend im Hintergrund laufen hatte, stehen ihre zufälligen Geräteschlüssel der letzten 14 Tage zum Hochladen zur Verfügung.
Auch ihren Positivkennungen werden die Übertragungsrisikograde VI, dreimal VIII, V, III und I rückwärts für die 7 vorausgegangenen Tage vergeben, allerdings beginnend mit dem 20. und damit einen Tag gegenüber Anton versetzt.
(Dass beide am selben Tag über ihr Testergebnis informiert wurden, weiß die CWA nicht. In der aktuellen Version steht ihr nur das Datum der Zustimmung zum Hochladen für die Ermittlung des tagesspezifischen Übertragungsrisikograds zur Verfügung.)
Für die 7 noch weiter zurückliegenden Tage, den Zeitraum vom 7. bis zum 13. des Monats, wird jeweils der Übertragungsrisikograd I vergeben.
||||||||||||||||
|-|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|:-:|
|Übertragungsrisikograd|VI|VIII|VIII|VIII|V|III|I|I|I|I|I|I|I|I|
|Abstand zum Tag der Upload-Zustimmung|1|2|3|4|5|6|7|8|9|10|11|12|13|14|
|Erzeugungsdatum des Schlüssels|20.|19.|18.|17.|16.|15.|14.|13.|12.|11.|10.|9.|8.|7.|
Tabelle 2: Übertragungsrisikograd für Aishas 14 geteilte Positivkennungen, basierend auf dem Abstand zum Tag der Upload-Zustimmung (21.)
Anton und Aisha fahren regelmäßig zusammen zur Arbeit.
Betty hat denselben Arbeitsweg und sitzt gelegentlich im selben Bus.
Alle drei beschäftigen sich während der Fahrt mit ihren Smartphones, so dass den Bluetooth-Signalen keine Hindernisse im Weg sind.
Betty hat Anton und Aisha am 9. und am 16. morgens und abends für jeweils 10 Minuten getroffen.
Anton saß dabei einen Meter von ihr entfernt und Aisha zwei.
Als Bettys CWA am 21. Antons Positivkennungen abruft und an die Schnittstelle des Betriebssystems ihres Smartphones übergibt, wird eine Begegnungsmenge für den 16. erkannt.
(Für den 9. hat Antons CWA keine Positivkennung hochgeladen.)
Diese Begegnungsmenge hat insgesamt 20 Minuten gedauert und die Smartphones waren im Durchschnitt einen Meter voneinander entfernt.
Daraus ergeben sich Werte von jeweils 1 für die Dauer und die Dämpfung (s.a. "Konfiguration von Begegnungen" im Abschnitt "Aktuelle Konfiguration").
Damit ist sichergestellt, dass diese Begegnungsmenge nicht verworfen wird.
Der Wert des Parameters für das Verzugsrisiko (21 - 16 = 5 Tage) ist durchgehend mit 5 konfiguriert und der Wert des Parameters für das Übertragungsrisiko wird eins-zu-eins vom Übertragungsrisikograd übernommen, beträgt also 8.
Das Begegnungsrisiko errechnet sich damit als 1 x 1 x 5 x 8 = 40, was übrigens der höchste mit der aktuellen Konfiguration erreichbare Wert ist.
Der Grenzwert von 11 wird überschritten, womit die Begegnungsmenge als Risikobegegnung zählt.
| | | | | | | | | |
|-|-|-|-|-|-|-|-|-|
|Verzugsrisiko| >=14d (5) | 12-13d (5) | 10-11d (5) | 8-9d (5) | 6-7d (5) | **4-5d (5)** | 2-3d (5) | 0-1d (5) |
|Dämpfung| >73dB (0) | >63-<=73dB (1) | >51-<=63dB (1) | **>33-<=51dB (1)** | >27-<=33dB (1) | >15-<=27dB (1) | >10-<=15dB (1) | <=10dB (1) |
|Dauer| 0min (0) | >0-<=5min (0) | >5-<=10min (0) | >10-<=15min (1) | **>15-<=20min (1)** | >20-<=25min (1) | >25-<=30min (1) | >30min (1) |
|Übertragungsrisiko| I (1) | II (2) | III (3) | IV (4) | V (5) | VI (6) | VII (7) | **VIII (8)** |
Tabelle 3: Parameterwerte für Bettys Begegnungsmenge mit Anton am 16.
Da diese Risikobegegnung zugleich die einzige Risikobegegnung ist, die Bettys CWA bekannt ist, wird auch nur sie bei der summarischen Auswertung ihrer Aufenthaltszeiten in den Entfernungsräumen bis 1,5 Meter und bis 3 Meter berücksichtigt.
Betty hielt sich 20 Minuten im Entfernungsraum unter 1,5 Meter auf, die zur Gänze gezählt werden.
Auch bei der Verrechnung dieser 20 Minuten mit dem Begegnungsrisiko der Risikobegegnung mit dem höchsten Risiko kann wieder nur diese eine Risikobegegnung mit ihrem Begegnungsrisiko von 40 berücksichtigt werden.
Die Multiplikation der 20 Minuten mit 40/25 (25 ist der aktuell konfigurierte Wert für "durchschnittlich riskante" Risikobegegnungen; s.a. *risk-score-normalization-divisor* in "Konfiguration von Dämpfung & Dauer" im Abschnitt "Aktuelle Konfiguration") ergibt 32 Minuten.
Da die CWA ab 15 Minuten über ein erhöhtes Risiko benachrichtigt, erhält Betty nun eine solche Benachrichtigung.
Zugleich wird ihr mitgeteilt, dass sie eine Risikobegegnung hatte und diese 5 Tage zurückliegt.
Am folgenden Tag, dem 22., ruft Bettys CWA auch Aishas Positivkennungen ab.
Sie erkennt zusätzliche Begegnungen am 16. und am 9. des Monats.
Beide Begegnungsmengen haben insgesamt jeweils 20 Minuten gedauert und die Smartphones waren im Durchschnitt zwei Meter voneinander entfernt.
Auch hieraus ergeben sich Werte von jeweils 1 für die Dauer und die Dämpfung.
Die Verzugsrisikowerte (22 - 16 = 6 Tage; 22 - 9 = 13 Tage) sind konstant 5 und die Übertragungsrisikowerte 5 für den 16. und 1 für den 9. des Monats.
Die Begegnungsrisiken berechnen sich für den 16. also als 1 x 1 x 5 x 5 = 25 und für den 9. als 1 x 1 x 5 x 1 = 5.
Die Begegnungsmenge des 9. erreicht den Grenzwert nicht und zählt damit nicht als Risikobegegnung.
|| | | | | | | | |
|-|-|-|-|-|-|-|-|-|
|Verzugsrisiko| >=14d (5) | 12-13d (5) | 10-11d (5) | 8-9d (5) | **6-7d (5)** | 4-5d (5) | 2-3d (5) | 0-1d (5) |
|Dämpfung| >73dB (0) | >63-<=73dB (1) | **>51-<=63dB (1)** | >33-<=51dB (1) | >27-<=33dB (1) | >15-<=27dB (1) | >10-<=15dB (1) | <=10dB (1) |
|Dauer| 0min (0) | >0-<=5min (0) | >5-<=10min (0) | >10-<=15min (1) | **>15-<=20min (1)** | >20-<=25min (1) | >25-<=30min (1) | >30min (1) |
|Übertragungsrisiko| I (1) | II (2) | III (3) | IV (4) | **V (5)** | VI (6) | VII (7) | VIII (8) |
Tabelle 4: Parameterwerte für Bettys Begegnungsmenge mit Aisha am 16.
|| | | | | | | | |
|-|-|-|-|-|-|-|-|-|
|Verzugsrisiko| >=14d (5) | **12-13d (5)** | 10-11d (5) | 8-9d (5) | 6-7d (5) | 4-5d (5) | 2-3d (5) | 0-1d (5) |
|Dämpfung| >73dB (0) | >63-<=73dB (1) | **>51-<=63dB (1)** | >33-<=51dB (1) | >27-<=33dB (1) | >15-<=27dB (1) | >10-<=15dB (1) | <=10dB (1) |
|Dauer| 0min (0) | >0-<=5min (0) | >5-<=10min (0) | >10-<=15min (1) | **>15-<=20min (1)** | >20-<=25min (1) | >25-<=30min (1) | >30min (1) |
|Übertragungsrisiko| **I (1)** | II (2) | III (3) | IV (4) | V (5) | VI (6) | VII (7) | VIII (8) |
Tabelle 5: Parameterwerte für Bettys Begegnungsmenge mit Aisha am 9.
Die Risikobegegnung des 16. wird hingegen bei der aktualisierten summarischen Auswertung mit berücksichtigt, so dass Bettys CWA nun 20 Minuten (mit Anton) im Entfernungsraum bis 1,5 Metern ganz und weitere 20 Minuten (mit Aisha) im Entfernungsraum bis 3 Meter zur Hälfte (also als 10 Minuten) zählt.
Die so ermittelten 30 Minuten werden wieder mit Bettys Risikobegegnung mit Anton verrechnet, die mit 40 nach wie vor das höchste Begegnungsrisiko aller erkannten Risikobegegnungen des aufgezeichneten 14-Tage-Zeitraums hat, so dass sich 30 x 40/25 = 48 Minuten ergeben.
Bettys aktualisierte Risikobenachrichtigung zeigt jetzt 2 Risikobegegnungen an, von denen die letzte 6 Tage zurückliegt.
## Aktuelle Konfiguration
Wie im [Abschnitt 'Risk Score Calculation'](https://github.com/corona-warn-app/cwa-documentation/blob/master/solution_architecture.md#risk-score-calculation) des Solution-Architecture-Dokuments beschrieben, werden die jeweilgen Parameter für die Berechnung von einer Menge an Parametern bestimmt, die von cwa-server zur Verfügung gestellt werden. Diese Konfiguration kann sich über die Zeit auf Basis der jüngsten Forschungsergebnisse ändern. Die jeweils aktuell gültigen Parameterwerte können im [cwa-server-Repository](https://github.com/corona-warn-app/cwa-server) eingesehen werden:
Wie im [Abschnitt "*Risk Score Calculation*"](https://github.com/corona-warn-app/cwa-documentation/blob/master/solution_architecture.md#risk-score-calculation) des *Solution-Architecture*-Dokuments beschrieben, werden die jeweiligen Parameter für die Berechnung aus von einer Menge an Parametern bestimmt, die vom CWA-Server zur Verfügung gestellt werden.
Diese Konfiguration kann sich über die Zeit auf Basis neuer Forschungsergebnisse und Erkentnisse ändern.
Die jeweils aktuell gültigen Parameterwerte können im [*CWA-Server-Repository*](https://github.com/corona-warn-app/cwa-server) eingesehen werden:
- [Konfiguration von Begegnungen](https://github.com/corona-warn-app/cwa-server/blob/master/services/distribution/src/main/resources/master-config/exposure-config.yaml)
- [Konfiguration von Dämpfung & Dauer](https://github.com/corona-warn-app/cwa-server/blob/master/services/distribution/src/main/resources/master-config/attenuation-duration.yaml)
- [App-Konfiguration, inkl. des minimalen Risikowerts](https://github.com/corona-warn-app/cwa-server/blob/master/services/distribution/src/main/resources/master-config/app-config.yaml)
- [Risikowertklassifizierung](https://github.com/corona-warn-app/cwa-server/blob/master/services/distribution/src/main/resources/master-config/risk-score-classification.yaml)
- [Risikowertklassifizierung](https://github.com/corona-warn-app/cwa-server/blob/master/services/distribution/src/main/resources/master-config/risk-score-classification.yaml)