Relative vs absolute risk: the trick behind scary headlines
"Doubles your risk" and "raises it from 1 in 10,000 to 2 in 10,000" can describe the exact same finding. One sells papers; the other tells you what to do.
Here is the most useful piece of arithmetic in all of health reporting, and it takes two minutes to learn.
Imagine a condition that affects 1 person in 10,000 each year. A study finds that a certain habit doubles the risk. Alarming, and true. But doubling 1 in 10,000 gives 2 in 10,000. The habit adds one case per ten thousand people. The very same fact can be written two ways:
- Relative risk: "increases the risk by 100%."
- Absolute risk: "raises it from 1 in 10,000 to 2 in 10,000."
Both are honest. Only one helps you decide anything.
Why the relative number travels further
Relative risk is bigger, rounder, and frightening, so it is what makes the headline. It is also what a press release reaches for when the underlying effect is real but tiny. "Cuts risk by a third" sounds like a reason to change your life. A third of what? If the risk was small to begin with, a third of it is smaller still.
The reverse trick exists too. Benefits are often sold in relative terms to look impressive, while harms get quoted in absolute terms to look negligible, or the other way round, depending on who is talking.
The one question
Whenever you meet a percentage change in risk, ask: from what, to what, in absolute numbers? A good study, and a good article, will tell you without being asked. When only the relative figure is on offer, treat the claim as unfinished.
This is why our reviews try to give the plain frequencies, so many in a hundred, or a thousand, rather than the percentage swing. It is less dramatic. It is also the only version you can actually use.
This is an evidence summary for the general reader — not a diagnosis or personalised medical advice. For any health decision, talk to a qualified professional.
More stories
How we weigh the evidence
Every claim we assess carries the same two marks: how strong the evidence is, and which way it points. This is the method behind them, and the reason we sometimes say "we don't know."
The evidence pyramid: why not all studies are equal
"A study found" tells you almost nothing until you know what kind of study it was. A quick tour from anecdote to meta-analysis.
Does vitamin C stop you catching colds?
Decades of Cochrane trials find no meaningful drop in cold risk for ordinary adults, only a modest shortening of colds once they start.