I have this dream. An educational resource to teach children to think: numeracy, logic, and how different an issue seems from an individual versus a population perspective. In part, this is my response to anti-vaccine arguments that flourished following a 1998 Lancet article; but are as old as the first vaccine:
In 1796, Edward Jenner demonstrated that prior ‘vaccination’ with cowpox would lead to protection against smallpox. Smallpox was one of the worst scourges of humanity, until eradicated in 1977 through a global immunization programme, led by the World Health Organization.
Anti-vaccine arguments date back to Jenner: like most vaccines smallpox can cause serious adverse reactions, but it does not turn anybody into a cow – as depicted in the first image of the anti-vaccine movement.
I don’t know if anybody seriously believed that smallpox vaccine – cowpox that has been altered through multiple generations of human transmission – could turn one into a cow. But the list of ills caused by modern vaccines, according to some anti-vaxxers, covers a large range of human illnesses for which we either have not yet defined the cause or there are multiple causes.
In fact, smallpox vaccine has amongst the worst safety profiles of vaccines. Luckily for humans, this is now only of historical interest. Smallpox was the first disease eradicated by a vaccine; though some anti-vaxxers can provide entertaining accounts of why it really has not been eradicated, or that its eradication was purely coincidental to the vaccine programme.
In 1999, I was invited to write WHO guidance on adverse events following immunisation. The AEFI is a carefully crafted concept that requires understanding of the logical fallacy of post hoc, ergo propter hoc. This is Latin for mis-attributing to coincidence a causal relation: when event A happens after event B, we can say that B caused A.
In reading through the science of AEFI, we classified events that happened after as being possibly, but not necessarily caused, by the vaccine. The event could be from an error in storing, preparing or giving the vaccine: a programme error. These remain too common in developing countries, the main audience for the WHO manual. Injection and anxiety related reactions are relatively common, especially in poorly planned mass campaigns. But coincidental events are the main problem for immunisation programmes are those that just happen to occur after immunisation. So, why is my claim for an event being coincidental stronger than the claim around smallpox eradication?
Of course, some AEFIs are caused by the vaccine. The known vaccine reactions have been well defined during initial trials; rarer events require wide-scale use in programmes to become evident. During my time leading the NZ immunisation programme, I found it ironic that the vaccine with the most severe known reaction (roughly 1 in a million risk of paralysis) was the one that caused the least public concern about vaccine risks. Perhaps because it was the only oral vaccine in the national immunisation schedule at that time.
In contrast, there is roughly a 1 in 200 risk of paralysis from infection with the poliovirus. The risk difference between disease and vaccine is why immunisation improves health. But human cognition is not well designed to compare risks. For example, we worry more about getting on an airplane than in a car; while on a tropical beach we fear sharks more than a coconut falling on our head. (I hope that you know the first risks are statistically much lower than latter.)
So, you can see that our intuitive risk meter is off; but we can correct this through deliberate process of logical thinking together with careful review of the available evidence. What I want is an education resource that helps children understand how to think in that deliberate process. Which involves understanding numbers and logic, as well as the inevitable flaws in all kinds of data. And the challenge of how to carefully appraise the evidence. So, what does this education resource look like?
I look forward to contribution, as I explore these ideas, starting with the number one and how 1+1 can equal 0,1,2 or 3 – and the intrinsic ‘slipperiness’ of numbers.