Needed: an Interdisciplinary Pandemic Modelling Platform
I agree with Alexander Kurz’s question above, that since a pandemic, such as Covid-19, is not only a biological phenomenon reducible to the SARS-COV-2 virus but extends to personal biological histories (what the individual immune system has learned from previous dealings with similar virii), personal hygienic practices, social practices (culture), geography, institutional fabric, information network (the internet makes a huge difference) etc… that one cannot simply apply a model from one country to another, without adding a lot of caveats. Nevertheless, this needed to be done to help test models given the sparse information available, and it was done by the Imperial College team in Report 13.
The Swedish strategy of controlled “herd immunity” is defended very clearly in the lengthy interview by Prof Johan Giesecke, advisor to the UN and Sweden.
He admits that their measures to protect the old failed. 50% of the deaths they had occurred in hospices, which in Sweden are much larger and centralised than the Norwegian ones, and also staffed by cheaper foreign labour that may not have understood the directives. (That was a hypothesis at the time, I guess they will be investigating). Norway’s hospices are much further apart and much smaller. So there is a major difference.
Prof Giesecke laments the UK’s change of strategic direction — which clearly made their life much more difficult in Sweden. His argument was that we will find that the Covid-19 epidemic was bad — flues are a lot worse than we civilians have been aware of — but not much worse than the normal seasonal flu. Perhaps 20% worse. The danger of the shutdown strategy is that, apart from being very bad for the economy, is that it will only delay the spread of the virus. He fears that countries who went into shutdown will then as soon as they re-open be confronted with a new spurt of growth in Covid-19 cases, leading to new shutdowns, new relaxation, new spurts of growth, … and that in the end, the same number of people will have died. Time will tell.
A question that should have been asked is given the difference between Norway and Sweden, if the cultural differences between countries may be what justifies different policies being applied in different localities. If so one may want future models to be much more fine-tuned to local cultural realities. This would constitute a reason not just for open source models, but for a large open-source modelling platform that would allow researchers from the humanities (eg. sociology) to cooperate in setting parameters to better model situations in different localities.
The economic impacts of the shutdown may seem like a secondary issue. But the case is made in the debate “The Cause of Death” (link) by Toby Young that the number of deaths worldwide related to economic hardship, could far outstrip the number saved by the measures, given for one, that close to one billion people are living near starvation levels. (It would be interesting to have more investigation in that direction to see how well-founded that argument is).
So this is clearly a highly complex issue with a huge number of interdependencies, all very difficult to model. That seems to demand an open interdisciplinary Pandemic modelling Platform, that could be used by researchers worldwide contributing from every field.
