Most alcohol-related harm doesn't come from alcoholics. It sounds wrong, and the instinct is to file it under clickbait, but the arithmetic holds. The heavy drinkers in the tail carry the highest individual risk, and yet the moderate-drinking middle is so much larger that it ends up accounting for around two-thirds of the total harm. I made a full video on why this week, with the sunburn analogy and the maths, so head to the channel if you want the walk-through.

In this issue of the newsletter I’ve asked myself what lesson we should learn from the prevention paradox, because I don't think it has a black & white answer.

When the epidemiologist Geoffrey Rose first described the paradox in 1981, the interesting part was never the arithmetic. It was the choice the arithmetic forces on you. You can run a high-risk strategy, where you go and find the people in the tail and treat them, or you can run a population strategy, where you nudge the whole curve a little to the left and accept that almost no single individual gets much back.

Rose's own conclusion leaned toward the population strategy, because that's the one that actually shifts the totals. In practice though, the high-risk strategy is often the one that feels right. It's targeted, it's matched to need, and the person you're helping is severe, identifiable and in front of you. The population strategy asks you to spend on a vast crowd of people who were mostly never going to get ill anyway, for a benefit so thin that no individual would thank you for it.

So here's my question, to which I honestly don’t have a perfect answer... We know, from the maths, that the middle carries most of the burden. So why does everything we do still pull us toward the tail?

Let’s start with our methods. The age-old saying is that “a QALY is a QALY is a QALY”. But it’s not really true anymore. Because NICE and many other HTA bodies around the world now have mechanisms in place that account for who is the recipient of those QALYs. Eventually, this leads us to concepts such as distributional cost-effectiveness analysis, but there are much simpler mechanisms already in place. Right now!

Take severity weighting for example, the kind NICE now applies via the 1.2 or 1.7 QALY modifier. This mechanism deliberately hands more value to gains in more severe disease. That isn't an accident or an oversight. It's the system formally encoding the same instinct, that the sickest have a stronger claim on our help. And precision medicine, for all its sophistication, rides that instinct too. We find the responders, target the mechanism, and concentrate the benefit on the people in the tail, where it can be valued cleanly and where it feels most deserved.

So the pull toward the severe isn't one bias in one place. It runs through our maths, through our value frameworks, and through the plain moral sense that you treat the sickest person in the room first.

This is where I have to be careful, because the easy move is to call all of this a mistake and walk off. I don't think it is.

"Help the most severe first" is not a cognitive error to be corrected away. It's a moral position, and a defensible one, held on purpose by a lot of thoughtful people. The severity modifier exists precisely because society decided a QALY gained by someone gravely ill is worth more than the same QALY gained by someone who is basically well. You can disagree with that, but you can't wave it away as irrationality.

So the prevention paradox doesn't actually tell us we're wrong to focus on the tail. What it does is far more uncomfortable. It tells us that focusing on the tail has a real, measurable cost, paid in all the cases out in the middle that we could have prevented and chose not to. That’s the underlying opportunity cost. The arithmetic doesn't settle the argument. It just refuses to let us pretend the trade-off isn't there.

So I'll leave it as a question rather than a verdict, because I think that's the best I can do.

When your next intervention reaches you, work out which of Rose's two strategies you go with. Are you aiming at the tail, or are you following the population approach? Just because Geoffrey Rose favoured one approach over the other, doesn’t mean that you have to!

Mirko

Full video on the Prevention Paradox is on my YouTube channel.

Where do you land on this? When the per-patient numbers favour the severe few but the population case favours the middle, which way should the call go, and which way does it actually go on your projects?

P.S. — Pharma Radar is written by Mirko von Hein. I help pharma and biotech teams navigate HTA submissions, cost-effectiveness modelling, and market access strategy across the UK, Ireland, and Germany. After a decade across IQVIA, Parexel, and Gilead, I'm now taking on selected engagements through Von Hein Consulting.