A cityscape of Manchester at night.

Precaution

Tom Forth, .

The huge success of the UK's vaccine development, procurement, and distribution program is enormously welcome. It will save thousands of lives in the UK, reduce the damage to our economy, and restore our freedoms more quickly. This success has been almost universally celebrated by politicians, journalists, scientists, and the general public in the UK. It is envied worldwide.

After a difficult five years since our vote to leave the EU, a gruelling year of pandemic lockdowns, and the death of well over 100 thousand of our compatriots, this is an opportunity that almost everyone has embraced to be unambiguously positive about the actions and decisions of the UK government. We can reasonably assume that almost anyone in the UK who is less than strongly positive about our vaccination program is not a fair judge of government performance.

Since many Britons have been consistently critical of the government on Brexit, and since the natural comparison for our actions is the EU and its members, this situation is a chance for people to prove themselves reasonable. This then lends credibility to their criticisms of how the government is handling other things.

Almost everyone is taking this opportunity. Many are taking it more enthusiastically than me, which is largely a good thing. I'm pretty miserable. You should be more positive than me. But I think that some people are going a little bit too far and letting their enthusiasm spill over. I don't think they realise, but it is scaring people abroad, often as much as they feel the actions of people abroad are scaring people here. Specifically I am referring to clinical trials, vaccine safety, and the decisions of the rest of the world on these matters, which are almost always very slightly more cautious than here in the UK.

How big is that number?

"YOU WHAT TOM?" you might be thinking. "VERY SLIGHTLY?!" you may rage. And indeed you might think that the USA and Switzerland not approving the AstraZeneca vaccine is much more cautious than the UK's rapid approval. You might think that the decision of the Norwegian, Icelandic, Thai, German and other medical regulators to suspend AstraZeneca vaccinations for a few days while they investigate a tiny number of reports of thrombosis following vaccination is much more cautious than the UK and Belgium continuing with vaccinations. You might think that the decision of most of the rest of the world to continue with the manufacturer-recommended dosing schedule rather than extending it is much more cautious than the UK's decision to extend the gap between vaccination doses to 12 weeks for all but the most vulnerable people.

I think that the UK is making the right decision. But this is a bit much.

You may think those things. I just disagree is all. You are in great company I should add, most UK explainers of statistics are on your side not mine. And this is a rare and pleasing situation where I really do understand your side. My emotions are on your side, they always have been. But my experience pushes me elsewhere.

Let me explain.

Precaution and ethics

A lot of commentators in the UK are talking about benefit to cost ratios as if this is how we decide how to develop, test, and use vaccines. It is not. We have agreed, in our democracies, and in debates among experts in ethics, and in public forums outside of democracies, that we don't develop and use vaccines like that. I've taken part in those debates, as a citizen and a scientist, and lost. Badly. And in most countries we've created quasi-independent government bodies that we expect to apply the cautious rules that we as a society set without much political interference.

If we optimised for benefit with Covid-19 vaccines we'd have taken the first vaccines designed by BioNTech in January 2020, injected them into half of a few thousand volunteers, deliberately exposed those people to the virus, and then observed the results. We'd have had our "is it safe? does it work?" answer within two months and saved millions of lives. Some of the volunteers may have died, this is part of the "cost" in the benefit to cost ratio. In a society of people like me, we'd have done that. But society is not, let's be glad, people like me.

When we chose not to do human challenge trials for Covid-19 vaccines in 2020 we moved away from maximising the benefit to cost ratio of our interventions. We moved a long way towards precaution and an ethical value judgement.

Our move in the direction of precaution and ethics is much more than just this one choice. We have made dozens of similar choices and we continue to make them and stand by them today even though we could change. For example, every country in the world is currently using all of their vaccines at the manufacturer-recommended dilution. This is despite the antibody assays conducted during trials suggesting that almost the same level of protection, a level of protection beyond our dreams a year ago, would be achieved at a dilution two to four times greater. With vaccine production certain to be the limiting factor in national vaccination efforts for the foreseeable future our choice not to dilute is a huge move away from optimising benefit and towards precaution.

Trolleys and trams

In trying to explain why they think that those medical regulators taking different decisions to the UK's medical regulator are wrong many UK explainers of statistics have taken to using one of the most widely-known thought experiements in ethics: the trolley problem. Should a person pull a lever to divert a tram from running over ten people at the cost of now deliberately killing one? What about at a ratio of a thousand to one? What if we're not even sure if there is one person on the diversion track?

Pushing the boundaries of utilitarianism is one of the most common debates in ethics. In this case it is valuable. But it is more much valuable when we consider that we have made far larger choices, and are sticking to those choices even as the far larger costs accumulate. Our tram, I'm sad to say, has already run over hundreds of thousands of people. This choice about suspending vaccinations for a few days is not a very consequential one, and that's before we get into the role of public trust in the independence of regulators in vaccine acceptance.

I think that the UK's medicine regulator and Belgium's medicine regulator have taken the right decision to continue using AstraZeneca vaccinations while reports of an absolutely tiny number of cases of thrombosis are investigated. We have chosen to divert the tram, and there probably isn't even a single person on the track to which we have diverted it. I also think that the UK's medicine regulator has made the right choice in extending the delay between vaccine doses.

But I am much less sure that other countries have got their decisions wrong, or that we should risk undermining public confidence in their independence and competence on what is such a small measure of precaution. The difference in the amount of precaution that is being applied by different regulators is tiny, almost invisibly tiny, compared to the precaution that every regulator has applied and that we continue to apply in many other areas, such as vaccine dilution.

One of the most important questions in public statistics is "how big is that number?". In this case, it is tiny.

blog comments powered by Disqus