21st June will be decided on data … and assumptions

Donald Rumsfeld famously (or perhaps infamously) talked about there being “known knowns, known unknowns and unknown unknowns.”  It’s another way of saying that you don’t know what you don’t know.  This is particularly true when we come to the decision the Government must make about COVID-19 unlocking by Monday 16th June.

What do we know? Well, we know that 55% of the population have been fully vaccinated against COVID-19 with about 75% on one dose. We know that with regards to the Delta variant, two doses of the vaccines provides significant protection against infection and even greater protection against serious illness and death. That’s the good news. On the negative side, we know that the number of cases has been rising rapidly and the Delta variant is now dominant in the UK.   We also know that this can still lead to hospitalisation.

However, the unknowns are around the question about whether the vaccine is effectively beginning to break the link between cases and hospital admissions.  In November/December about 9% of cases led to hospitalisation whereas evidence suggests that this has halved to about 5%. However, there are some assumptions behind that figure and therefore we have some “unknowns” to deal with at the moment.

The crucial assumptions we are working with are first that hospitalisation correlates to a more serious form of the illness. The second assumption is that there is a lag factor of about 10 days between a positive test being recorded and a hospital admission being recorded. 

But what if those assumptions proved to be wrong?  We are hearing reports from the North-West that there are fewer examples of serious illness amongst those admitted to hospital and anecdotal evidence suggests that what is happening is that cases are mainly amongst younger people who are admitted for precautionary reasons for shorter periods of time. If that is so, then hospitalisations are less likely to lead to people being put on mechanical ventilation and admitted to ICU.

What is unknown at this stage is whether these people would have been admitted during the first and second wave. In other words, is it that we are still seeing some of the types of hospitalisatino but not the severest, or is it that those going into hospital now would not have met the triage threshold for admission in 2020. This is important to know because, first of all, it raises questions ans to whether or not hospital is the best place for them to receive treatment. Secondly it means that one solution to rising hospital cases could be to reintroduce tighter triage thresholds.

However, there is also a third potential implication. If hospital capacity is significantly higher than at the height of the virus then as well as this leading to people being admitted who would not have been previously, it is also possible that they are being admitted earlier because the queue will have been significantly reduced.  If this is so, then the lag time between testing positive and being admitted to hospital could be much shorter.  If the lag time is between 10 days and 1 day then the percentage of cases admitted could be anywhere between 5% and 3%. 

If we have seen the rate of admissions reduce from 5% to 3% (with potential to reduce further over the next two or three weeks) then that completely changes the picture in terms of the impact of further re-opening.

Now, I am not arguing here that this definitely is the case. The original assumptions may still hold in which case it will make complete sense to delay June 21st. However, these are the sorts of questions that should be asked by the Government as they seek to make a judgement call. Furthermore, these are questions that depend on data not currently in the public domain.  In other words, we are dealing in “unknowns.”

This means that if the assumptions are wrong then despite seeking to follow the data, the Government may make the wrong decision. Furthermore, if they are wrong then we may still not know from the data in 2 -3 weeks time. If the assumptions are right then we are facing exponential increase in hospitalisations. If they are wrong then having the next tranche of over 50s and over 40s fully vaccinated means that at the peak of the 3rd wave, we will be looking at hugely reduced hospital figures. If the assumptions are wrong, then the solution to any increased hospitalisations among 20s -30s may be improved advice, triage and care at home.

This should remind us that whilst the mantra “data not dates” is partially true, it is not completely true.  Decisions are based on data analysed on the basis of assumptions. That’s why it is crucial that we challenge and check assumptions. I hope that the Government will use any delay not just to monitor the data and get out more vaccines but to also revisit their assumptions and check they have all the right data and the right assumptions for their models.

Most of us don’t have to get involved in making decisions about COVID but in our daily life and church life, we are often faced with important choices to make. It is important that we remember that our decisions depend both on known facts and assumptions. That’s okay as long as we know what the assumptions are and are willing to have them challenged and checked.

%d bloggers like this: