control (A) and Ethics (B)
Enforcement (A) and Informed consent (B)
Risk perceptions (A) and Accurate comparisons (B)
Central commands (A) and Public consultation (B)
Propaganda (A) and Balanced information (B)
Paternalism (A) and Person-centred care (B)
Obedience (A) and Critical thinking (B)
Scaremongering journalism (A) and Investigative journalism (B)
Older adults (A) and Young people (B)
Protecting life (A) and Living a worthwhile life (B)
Firefighting a zoonotic virus (A) and Animal welfare (B)
This is a formidable array of oppositions, but they should not be set against one another. In a well-balanced society they should blend in some sort of harmony. There is no perfect equilibrium of course. Different cultures afford different weights to each pair in normal circumstances. But when the scales are slanted so far to the left, something is surely amiss and critical reflection is sorely needed.
Interestingly, these two dualities seem anomalous:
Guesswork (A) and Evidence-based medicine (B)
Association (A) and Causation (B)
Strictly speaking, a perfect set of patterns would have these reversed, since all the other As appeal to supposed certainties. However, the pattern represents what has been emphasised (A) and what needs to be rebalanced (B), and there is little doubt that these two As are dominant.
4 Certainty and Uncertainty
Certainty (A) and Uncertainty (B)
The most striking general feature of my research into the pandemic is that it is difficult, if not impossible, to gain a clear view of what is happening. Government and health organisation predictions and announcements change daily: the World Health Organisation (WHO), for example, said for months that there was no evidence that healthy people should wear facemasks, and then changed that advice on June 6th. New information emerges in a bewildering stream, much of it in fragments posted by news-media searching for the latest story. Drawing watertight conclusions from the cascade of words and data is impossible even for ‘the experts'.
It doesn't matter which field their expertise is in. When one ‘expert’ makes an assertion, the only certainty is that a different expert will disagree. In striking contrast to this ambiguity, and despite published minutes of their own meetings displaying enormous doubt, scientists and politicians have made drastic consensus decisions ‘in the public interest'. One might say that this is their role, but decisions to act so decisively on barely any evidence require extensive scrutiny.
Decisive Planning in Colossal Uncertainty
Here, for example, are edited minutes of a meeting of the Scientific Pandemic Influenza Group on Modelling (SPI-M). According to the government's website the SPI-M gives expert advice to the Department of Health and Social Care and wider UK government on scientific matters relating to the UK's response to an influenza pandemic (or other emerging human infectious disease threats). The advice is based on infectious disease modelling and epidemiology, and the group members have a background in public health science and closely related disciplines (19). The minutes of other SPI group meetings can be found here (20).
On February 19th, the group considered the closure of schools in the UK. They said:
At this stage the magnitude of the impact that school closures would have on a UK epidemic of COVID-19 is very uncertain. There are many uncertainties about the virus including, importantly, the role of children in transmission, and the severity of infections in children … we assumed that children have a role in transmission similar to that of influenza…
Three new (models) were produced to address this… one full spatial individual-based model (IBM) exploring a range of scenarios, one individual-based model that considered also reactive closure strategies and varying infectivity profiles and one compartmental model exploring two different forms of age-structured mixing. The different modelling approaches gave results which were similar in some ways but differed in others…
Any impact from school closures on the total number of cases is likely to be highly limited. … The effectiveness of school closures in reducing peak incidence is sensitive to the reproduction number R0: the higher the R0, the less effective they would be…
The key difference between our approaches so far is the predicted reduction in the peak size of the epidemic that school closures produce. … For R0 around range 1.9–2.3 and school closures of 6–12 weeks: the IBMs suggests an effect of the order of 20–60% reduction, while the compartmental models suggest 7.5%–30%. (19)
(Note: the basic reproduction number (R0) of an infection is the expected number of cases directly generated by one case in a population. In commonly used infection models, when R0 is more than 1, the infection could start spreading in a population, but not if R0 is less than 1. It is assumed that the larger the value of R0 the harder it is to control the epidemic.)
On one model the reduction in ‘peak size’ could be as little as 7.5%. On another it could be as much as 60%. Which raises an obvious question: does it make sense even to discuss policy based on such wildly divergent guesses?
To be fair, given the number of variables and ‘what ifs', the modelling challenge is considerable:
Not all parameters and assumptions have yet been cross calibrated between models. Only national-scale closure policies have been compared so far. The knock-on effects on how contact patterns will change with school closures of different lengths is a fundamental knowledge gap that cannot be determined by modelling. … Detailed forecasts of the likely impact of school closures will be possible once there has been several weeks of sustained transmission of COVID-19 within the UK. (19)
In other words, we simply do not know anything like enough even to make a barely informed estimate of the effect of closing schools on the virus. Nor do we know nearly enough to judge the potentially massive effects on mental health, education and domestic stress that closing schools causes.
There were attempts to acquire firmer evidence, but this was not forthcoming (21):
Emergency school closures are often used as public health interventions during infectious disease outbreaks in an attempt to minimise the spread of infection. However, if children continue to mix with others outside the home during the closures, these measures are unlikely to be effective.
We searched four databases from inception to February 2020 for relevant literature. … Activities and social contacts appeared to decrease during closures but contact was still common. All studies reported children leaving the house or being looked after by non-household members. There was some evidence that older child age and parental disagreement with closure were predictive of children leaving the house, and mixed evidence regarding the relationship between infection status and leaving the home. Parental agreement with closure was generally high, but some parents disagreed due to
