I’ve ranted several times already about publication and other biases that permeate the social (and some other) sciences without these disciplines taking much serious notice. Like with some of my other fancies, after a couple of such posted outbursts, I keep coming across many potential triggers, confirmation bias influence is inescapable, but habituation sets in, the triggers just don’t evoke the visceral upset anymore that is required to get me writing. Going on and on about the same subject seems a bit silly, venting the same opinion in slightly different phrasings time and again makes me feel like a player on repeat. Unfortunate, because hedgehogs rule the waves, but one has no say in such things (never mind the self-help industry).
So that this guy manages to get my juices flowing again is telling:
And it is especially telling because I don’t think this is a particularly good presentation. But David Newman, part of, and spokesperson for this particular evidence-based medicine initiative, does two notable things in here. He not only shows the extremely distorted way scientific info is funneled to us, the thing that normally gets at me. He also promotes a very simple descriptive statistic, the Number of people Needed to Treat for one person to benefit (NNT), that makes understanding what a medical treatment contributes to your health. That’s clever. For those intrigued, have a go at this interesting article in Wired about him.
But I intuit that the stat as intended wouldn’t have been sufficient trigger for me to devote another post to this fancy. What this stat did to me however, was less related to its power to indicate the effectiveness of a treatment, as to its power to convey its ineffectiveness. Yes, some very simple things, that require knowing about them, and/or access to them, like basic hygiene, clean drinking water, and some other non-pharmaceutical ‘interventions’, do save huge numbers. Huge numbers in terms of proportions of total populations. Beyond that, what are we offered as ‘treatments’? The example used in the presentation, the Mediterranean ‘diet’, the most powerful coronary intervention available, benefits 1-in 30 high-risk individuals (translate that into prevents one additional death for every 30). That means that 29 out of 30 live or die, dining Mediterranean or not. We are not the arbitrators of who lives and who dies as Newman phrases it. A sobering thought.
One that may linger and turn sour, so let me remind you of something that is undoubtedly saving lives, the simple act of spreading some enjoyment into otherwise pretty drab environments: