Core Revision
Revision aid, not clinical advice. This page is educational material written for exam preparation. It is not a substitute for the current guideline or for clinical judgement, and must not be relied on for the care of any individual. Always check up-to-date national guidance and verify independently before acting on anything here.
Relative risk, absolute risk, and the number the AKT wants you to spot
Dr Victoria HammondCo-Founder & Lead for Educational Design · 4 min read
Published 31 May 2026 · Checked against sources 28 September 2026
Statistics questions are some of the most reliable marks in the whole AKT — not because they’re easy, but because the same handful of tricks come up again and again. Learn to recognise them and you’re answering pattern, not maths. The biggest of them all is the gap between relative and absolute risk.
Drug companies, headlines and exam writers all love relative risk for the same reason: it sounds dramatic. “Halves your risk of a heart attack!” is a relative statement, and it can be technically true while the actual benefit to any individual patient is tiny. The AKT tests whether you can see through that — so let’s make the distinction concrete.
Relative risk reduction tells you the proportion of risk removed. Absolute risk reduction (ARR) tells you the actual percentage-point difference for a real person. The same trial can show a huge relative reduction and a trivial absolute one — and only the absolute figure tells a patient what the treatment does for them.
One worked example, and you’ll never unsee it
Imagine a trial of a new drug to prevent heart attacks over five years.
| Group | Heart attacks over 5 years |
|---|---|
| No drug (control) | 2 in 100 (2%) |
| New drug | 1 in 100 (1%) |
Relative risk reduction
The risk fell from 2% to 1% — that’s a drop of one half. So the relative risk reduction is 50%. This is the number that makes the headline.
Absolute risk reduction
But in real terms, risk fell from 2 in 100 to 1 in 100 — a difference of just 1 percentage point. That’s the absolute risk reduction: 2% − 1% = 1%.
Number needed to treat (NNT)
NNT is simply 1 ÷ ARR (as a decimal): 1 ÷ 0.01 = 100. You’d treat 100 people for five years to prevent one heart attack. Same trial, same drug — “50% reduction” and “NNT of 100” are both true, and they feel very different.
- “By how much?” usually wants the absolute figure. If a stem quotes an impressive percentage, check whether it’s relative — the trap answer treats a relative reduction as if it were absolute.
- NNT = 1 ÷ ARR. Memorise that one formula; many “calculate the NNT” items are a single division once you’ve found the ARR.
- Lower NNT = more effective (fewer people treated per benefit). NNH (number needed to harm) is the same maths for side effects — a high NNH is reassuring.
- Communication items: the “best” way to explain a treatment to a patient is almost always in absolute terms or a natural-frequency statement (“1 in 100”), never the relative percentage.
Sitting the AKT? Try twenty questions free.
Twenty original AKT questions in the exam's 80/10/10 split: sixteen clinical, two on evidence-based practice and two on organisation and management, with the reasoning after every one and a marked report. Free, no card and the first five are open to all.
The patient-facing version — which the SCA tests too
This isn’t only an AKT party trick; it’s good consulting. A patient deciding whether to start a statin deserves “this lowers your chance of a heart attack over the next ten years from about 1 in 10 to about 1 in 13,” not “it cuts your risk by 25%.” Natural frequencies with a consistent denominator are the format people actually understand — and the exams reward you for using them.
- A big relative number with no baseline risk. “Doubles the risk” is meaningless until you know whether the baseline is 1 in a million or 1 in 10.
- Confusing ARR with relative. If you can subtract the two event rates directly, that’s the absolute reduction — don’t divide.
- Forgetting the time frame. An NNT is meaningless without “over how long”. Always read the duration in the stem.
When a statistics stem quotes a percentage that sounds impressive, my first move is always the same: find the two actual event rates and subtract them. That single habit — convert everything back to “X in 100, before and after” — defuses almost every relative-risk trap the AKT can throw, and it’s exactly the number your patient needed anyway.
Dr David Palethorpe · MBBS, BSc (Hons), MRCGP · interests include medical statistics & EBM
Want the full picture? This is one slice of the statistics tested in the AKT. Our free AKT Statistics Guide walks through sensitivity and specificity, predictive values, confidence intervals and more, with worked examples — no sign-up needed. A revision aid, not a substitute for the full curriculum.
