one plus one does not always equal two

I don’t know enough about the specific case here to make a judgement, but this line leapt out at me:

as judged by QALYs, stopping 5400 people from getting a short viral illness from which all recover is more worthwhile than stopping one 3-month-old dying

[QALY = quality-adjusted life years]

So, if QALYs can’t simply be summed, how should they be combined to make a judgement?

---

For all my social networking posts, see my Google+ page