By Howard White, CEO, The Campbell Collaboration
How much is a life worth? At the recent Symposium on Systematic reviews organized by KU Leuven, Marijke Eyssen discussed a case when the Belgian Health Care Knowledge Centre (KCE) took an explicit stand on this issue.
An intervention for people at risk of heart attacks was found to be effective, raising life expectancy from around six months to just over four years. But it was expensive: one quarter of a million euros per patient, equal to €82,000 per additional Quality Adjusted Life Year. KCE recommended the intervention should not be adopted on cost grounds. The press didn’t like this: ‘KCE says your life is worth less than €82,000’.
A member of the audience suggested that valuing life is a political decision that should be left to politicians. And, indeed, politicians take decisions all the time which implicitly value life, since many interventions save lives or cause deaths. But they don’t explicitly put a value on life. It is not a vote-winner. A politician is far more likely to say ‘you can’t put a value on life’. This is a nonsense statement. Politicians do it all the time, just not explicitly.
This very same discussion had taken place at another meeting two weeks earlier, organized by the European Commission’s Joint Research Centre in a discussion on teaching critical evidence appraisal to policy makers, journalists and the general public. I thought this discussion would be about distinguishing counterfactuals from correlations. But another participant immediately pointed out that most people rely on far less reliable sources of ‘evidence’ such as anecdote, sentiment, personal experience and what they hear from some bloke they meet in the pub.
‘Intuitive evidence’ results in irrational decisions. We need to confront irrationality with the facts of what an evidence-informed decision would look like.
For example, consider all the irrational decisions we all make from a relative risk perspective. The amount of time and money spent on safety for different modes of transport is inversely proportional to the risk they pose. Mile for mile, air travel is safest, followed by trains and then cars. But we don’t get a safety demo every time we turn on the ignition.
A topical example is death by terrorist attack. Whilst these are terrible events, the amounts spent on preventing them is disproportionate compared to the risk they pose to us. A recent article in Business Insider listed the likelihood of other forms of death compared to a terrorist attack by refugee terrorists: the average American is 6 times more likely to die from a shark attack (one of the rarest forms of death on Earth), 29 times more likely to die from a regional asteroid strike, 260 times more likely to be struck and killed by lightning, 4,700 times more likely to die in an airplane or spaceship accident, 129,000 times more likely to die in a gun assault, 407,000 times more likely to die in a motor vehicle incident and 6.9 million times more likely to die from cancer or heart disease. But, like putting a value on life, a politician is more likely to say ‘Any risk is too high’. This is of course another nonsense statement, as we are taking a risk every time we step out of our front door. Or, indeed, when we don’t - since many more common causes of death, such as choking on food, might take place in the home.
So, before we teach critical evidence appraisal, we need first to confront irrationality. The starting point is a proper understanding of some basic principles, e.g. that there is no zero risk option, and that life cannot have unlimited value because resources are limited. We should also understand basic concepts – such as probability, relative risk and the value of quality of life.
In practical terms, researchers should engage policy makers in a two-way process. Researchers need to understand the context in which policy makers operate, to discuss what evidence is relevant, appropriate and usable. These discussions are part of a process of relationship-building, which I will take up in my next blog on the soft skills needed for evidence-based policy and practice.