The European experience - watching for transport-induced clusters

[whole diary]

[grahame]

Many European countries are "ahead of the curve" of the UK^▸ (United Kingdom) coming out of lockdown and tracking new infection clusters - so it's probably in the UK's interest to watch the continental experience, even if ("ssshhh") we don't want to be part of Europe.   A contact writes:

During discussion at  [snip] meeting on the effects of Covid-19 on public transport use our attention was drawn to reports that in some EU^» (European Union - about) states there appears to be a lack of evidence that any cluster of cases can be traced back to visits to business premises or the use of public transport.  I subsequently received links to descriptions of the research outcomes.  These can be found, for Austria, at https://www.ages.at/service/service-presse/pressemeldungen/epidemiologische-abklaerung-am-beispiel-covid-19/ and for France at https://www.leparisien.fr/societe/coronavirus-pourquoi-aucun-cluster-n-a-ete-detecte-dans-les-transports-05-06-2020-8330415.php

I am double at a loss underling neither the language nor the science ... but I know we have one or two members who are a step or a leap ahead of me.   

Google Translate from French suggests

The fears were great, but the situation seems rather reassuring. None of the 150 “clusters” detected between May 9 and June 3 (including 96 still considered “active”) are located in transportation, according to the latest weekly epidemiological bulletin posted online on June 4 by Public Health France.

Assistance in reading and evaluating the reports would be useful - anyone? 

[stuving]

The article in Le Parisien, points out:

To be picked up as a contact, an interaction has to fit the public health definition - close enough (presumably within 1 m) and for long enough (usually 15 minutes). if not, And to be a cluster, there must be at least three victims in one "place" - though extended to the same service on seven days in a row for trains.

Longer-distance travel was only allowed from a few days ago, so the results quoted are mostly for shorter trips, almost entirely to go to work. People don't talk to each other often in such cases, they keep a apart. And talking is, after coughing and sneezing, the biggest generator postillions (droplets).

Note that the StopCovid app only started a few days ago, so won't have had any input to the figures.

I would also point out that while France (or rather their version of SAGE) has declared the pandemic "under control", there are still 1000 new cases found per day. Very few of those are in clusters.

[JontyMort]

And talking is, after coughing and sneezing, the biggest generator of postillions (droplets).

...in which case it would - for once - presumably be useful for them to be struck by lightning?

It’s actually “postillons” - a shame, really. 🙂

[stuving]

And talking is, after coughing and sneezing, the biggest generator of postillions (droplets).

...in which case it would - for once - presumably be useful for them to be struck by lightning?

It’s actually “postillons” - a shame, really. 🙂

It must be morphic resonance, if you remember that ...

[stuving]

To be picked up as a contact, an interaction has to fit the public health definition - close enough (presumably within 1 m) and for long enough (usually 15 minutes). if not, And to be a cluster, there must be at least three victims in one "place" - though extended to the same service on seven days in a row for trains.

I've realised that's not what was meant. The seven days is, of course, the period during which the three cases are detected, and the train (or bus etc.) is s single service.

[grahame]

Some interesting stuff here:

Coronavirus: what if we stop repeating that public transport is risky?

https://medium.com/@alejandro.tirachini/coronavirus-what-if-we-stop-repeating-that-public-transport-is-risky-5ae30ef26414

The cost of allowing public transport´s free fall due to the COVID-19 pandemic, or reducing public transport to serve only those who have no choice, is immeasurably greater than the cost of taking all possible measures today to make public transport use safe for the population and for those who work in public transport. The dream of public transport being a motor of social integration rather than of social segregation is today more distant than ever. With the COVID-19 pandemic, people have abandoned public transport but not uniformly: high-income groups have left public transport in larger numbers.

What is a “reasonable” occupancy level for a public transport vehicle or station in the COVID-19 era? Being on a normal 12-meter bus with 10 or 15 passengers is more dangerous than entering a supermarket, a pharmacy or a public office? And what if the bus has 20 or 30 passengers? What is the difference if all passengers wear high-quality masks versus if only some passengers do it?

Could it be that at the end of this mess, we will discover that car trips are as or more responsible for the spread of the COVID-19 virus than public transport trips? We do not know yet.

Here it hurts repeating ad-infinitum “avoid using public transport”, instead of providing the conditions for travel on public transport to be effectively safe, and also be perceived as safe by its users. Associating public transport indefinitely with the spread of coronavirus will condemn it to be used only by those who have no choice.

[JontyMort]

To be picked up as a contact, an interaction has to fit the public health definition - close enough (presumably within 1 m) and for long enough (usually 15 minutes). if not, And to be a cluster, there must be at least three victims in one "place" - though extended to the same service on seven days in a row for trains.

I've realised that's not what was meant. The seven days is, of course, the period during which the three cases are detected, and the train (or bus etc.) is s single service.

Yes, it’s “...when at least three confirmed or probable cases, [who all] travelled in the same aircraft/train/bus/boat etc, are detected within a period of seven days”. The seven days is the period of detection, rathe than travel.

[4064ReadingAbbey]

I’ve translated the first part of the Austrian paper to give a feel for its contents.

Situation in Austria

Last changed 2 June 2020

Epidemiological clarification is about showing how an outbreak spreads within the population: To do this, one tries to identify sources of the infection or chain of transmission of the cases through personal interviews with sick or positively tested people (= cases). Knowing how the disease spreads among the population means that steps can be taken that are most likely to help curb or slow the spread.

An epidemiological investigation is not static: the number of cases investigated and their assignment to clusters change with the progress of the epidemiological investigation. For the assignment to a cluster, the setting is selected in which most of the transfers took place within the respective case cluster.

Table 1: Cluster analysis, as of May 29.
Cluster cases 5,357
Cluster 355
Cluster types 5
Cluster settings 12

On 29.05.2020 5,357 of a total of 16,581 COVID-19 cases could be assigned to one of 355 clusters (see Table 1). The clusters are currently divided into 5 cluster types. There are so-called cluster settings in each cluster type: This means that the majority of infections (= cluster case) are based on certain settings, e.g., leisure activity, family, work or retirement home and can be traced. But there are also settings with more than one relevant setting for the transfer.

The Cluster types shown in Figure 1 are:

• In tourist groups
• In tourist groups and further infections locally
• Travel-associated and further infections locally
• Contact with foreign tourists and further infections locally
• local cluster.

There is a lot more detailed analysis but no reference to public transport specifically.  However the article makes the following observation, one out of a list of bullet points:


Transmission occurs when several people are in close contact at the same place for a long time (cumulatively for 15 minutes, e.g. 1 person in close proximity for 15 minutes or 3 people in close proximity for 5 minutes).