The scientific establishment in this country has had a bad war. Its mistakes have probably made the Covid-19 epidemic, as well as the economic downturn, worse. Britain entered the pandemic late, with lots of warning, so we should have done better than other countries. Instead we are one of the worst affected in Europe and one of the last to begin to recover.
Not all the mistakes were driven by science. The decisions by Public Health England not to go out to the market for testing, protective equipment and logistics, to cease testing almost completely in March and to send people to care homes from hospitals affected by the virus – these were just bureaucratic bone-headedness. But the obsession with mathematical modelling lies behind other mistakes and continues to this day with the ridiculous fixation on a meaningless generalisation called R.
Expecting a typical flu virus, scientists were surprised by the explosion of cases in hospitals and assumed it signified exponential take-off of the virus in the community. In fact it was mainly the rapid spread of infection within hospitals, some of which probably started with staff returning from holiday in Italy and Spain. It was known in February that the virus is dangerous to the elderly and ill yet could be spread by the young and healthy. Was it therefore not obvious that infection control in hospital wards and care homes was vital? Apparently not if modelling an epidemic in a homogeneous “community” is your guide.
Government advisers became over-reliant on models that were both too complex and too simplistic at the same time, and failed to challenge underlying assumptions. The Imperial College model does not take into account high variability in social connectedness or susceptibility to infection among otherwise similar people. We now know that about 10pc of “superspreaders” cause 80% of infections, primarily because they meet many more people – which also makes them much more likely to become infected.
If you tell the models there is thus a correlation between susceptibility and infectiousness you get much lower forecasts of cases and deaths. Add that we now know that cross-immunity from common colds probably allows 40-60pc of the population to resist Covid-19, and the result is – as the work of Gabriela Gomes at the Liverpool School of Tropical Medicine indicates — that herd immunity is probably reached when as little as 15pc of the population is infected, rather than the 50-60pc implied by Imperial’s model. Hence the epidemic is petering out in London despite crowded streets.
True, not all of this was known at the start, but that is exactly why it was a mistake to rely so heavily on models that were bound to have wrong assumptions. Modellers often come to mistake their models for the real world. There was an embarrassing moment in a press conference in April when a scientist was asked whether we could learn anything from Germany’s relative success in containing the epidemic and replied that our contribution was to lead the world in modelling.
We now know that outside the healthcare system the growth of the epidemic had ceased to be exponential before the lockdown. The peak of 3-day average deaths on 10 April implies that the peak of infections occurred around 20 March, before the country locked down, according to Professor Simon Wood of Bristol University. He argues that bans on large gatherings and voluntary social distancing would have been sufficient. The head of the Norway’s Institute of Public Health now says that the country’s lockdown was unnecessary.
The outcome of the epidemic in different countries or American states is pretty much uncorrelated with the severity of lockdown. Sweden, with no lockdown, did no worse than Britain and far better than the models predicted. By now the models say it should have had up to 40,000 deaths with a lockdown; it has had under 5,000 without. Had Sweden managed to keep the virus out of care homes and hospitals, as Germany partly did, it would have done much better than us despite no lockdown.
Reversing these mistakes will not be easy. Britain needs to get out of lockdown quickly, ditching the stable-door-locking policies like 2-metre distancing and travel quarantine before damage to the economy becomes terminal. And science needs to rethink its affair with models rather than data.