Published on 2026-07-25
Your risk 'doubles': doubles from what, exactly?
A relative risk like 'doubles' or 'triples' says nothing about the real odds until you know the starting number. Here's the question that fixes it.
Two headlines, same word. 'This habit doubles your risk of the disease' could mean your odds went from 40 percent to 80 percent. It could also mean they went from 1 in a million to 2 in a million. Both are, technically, a doubling. Only one of them should keep you up at night, and the headline rarely tells you which.
That gap has a name: relative risk against absolute risk. Absolute risk is your actual chance of something happening, the plain number. Relative risk only compares two groups to each other, and it can look enormous while the underlying reality barely moves at all. Cancer Research UK, in a piece written specifically to help readers decode scary health stories, walks through exactly this trap using real reporting.
A tripled risk that adds up to almost nothing
One story they examine claimed that CT scans in childhood can triple the chance of developing brain cancer later in life. Tripling sounds like a reason to panic. But according to Cancer Research UK's breakdown, the absolute risk behind that tripling amounts to one additional case of brain cancer for roughly every 10,000 children given the scan. The relative jump is real. The number of children actually affected is tiny, because the starting risk was already tiny.
Their piece flips the pattern around, too. A study linking daily alcohol to a '5 percent increase' in breast cancer risk sounds mild next to a tripling. Yet Cancer Research UK notes that a 5 percent relative rise translates to roughly 60 extra women in every 10,000 developing breast cancer, because breast cancer is common to begin with. A small relative number, applied to something that already happens a lot, can move more real people than a huge relative number applied to something rare. The size of the relative jump and the size of its real world effect are two separate questions, and a headline only ever answers one of them.
The same trick, run in reverse
Cancer Research UK also cites a report, following the Fukushima nuclear accident, that described cancer risk in an exposed group rising by 70 percent. Framed like that, it reads as an emergency. The absolute figures behind it, though, moved from 0.77 percent to 1.29 percent, an increase of roughly half a percentage point. Both descriptions describe the exact same underlying change. One is written to alarm. The other is written to inform.
'Doubled' is a fact about the ratio. It says nothing yet about the size.
When the scare became policy, briefly
In October 1995, the UK's medicines regulator warned that certain contraceptive pills increased the risk of a dangerous blood clot compared with older versions, a warning that made front pages across the country almost overnight. A later review of the episode, published in the journal Human Reproduction, found that the studies behind the alarm reported odds ratios between roughly 1.5 and 2.3 for the newer pills against the older ones, and argued that bias in how the data was gathered was a more likely explanation than a true difference in risk. The review also found that the actual absolute rate of clots among users of the newer pills was lower than the rate seen among users of the older pills only a few years earlier.
None of that nuance made it into most of the coverage at the time. Large numbers of women stopped taking the pill or switched brands within weeks, and the same review recorded a rise in unplanned pregnancies and therapeutic abortions in the countries where the scare landed hardest. Two years on, there was no epidemic of blood clots to justify the reaction. There was, however, a very real and measurable cost, paid because a relative figure travelled without the absolute one attached to it.
The one question worth carrying
None of this means relative risk is a dishonest measure. It has a real, legitimate use: comparing how one group fares against another, especially for researchers tracking whether a treatment or an exposure makes a genuine difference. The trouble starts the moment it travels alone, dressed up as the whole story, with the absolute number quietly left out of the sentence. So when a headline says a habit, a product or a place doubles, triples or dramatically raises a risk, hold that number lightly for one more second and ask the plainer question sitting underneath it: out of how many people, and how many of them, exactly? Most of the time, that single question turns a frightening headline back into a fact you can actually use, one way or the other.
Sources
- Absolute versus relative risk: making sense of media stories (Cancer Research UK, Cancer News) (news.cancerresearchuk.org)
- The 1995 pill scare revisited: anatomy of a non-epidemic (Human Reproduction, PubMed) (pubmed.ncbi.nlm.nih.gov)