Epidemiology · Effect magnitude
Risk & Effect Size Calculator
From a 2×2 of exposure and outcome: relative risk, odds ratio, the absolute risk difference, and the number needed to treat — each with a confidence interval, and the relative-vs-absolute contrast spelled out so a scary headline figure can't hide a tiny real change. Plus Cohen's d for continuous outcomes.
Counts of people with and without the outcome, in each group.
Result
In plain English
This turns a 2×2 table into the numbers behind health headlines — and deliberately shows the absolute change next to the scary-sounding relative one.
- relative risk (RR)
- How many times more (or less) likely the outcome is in one group than the other. “Doubles your risk” means an RR of 2.
- absolute risk difference
- The actual change in your chances, in plain percentage points. “2 in a million instead of 1” is a doubling, but a tiny absolute change.
- odds ratio (OR)
- A close cousin of relative risk used in case-control studies. It overstates the RR when the outcome is common.
- NNT
- Number Needed to Treat: how many people you'd have to treat for one extra person to benefit.
- Cohen's d
- For continuous outcomes (not yes/no): how far apart two groups' averages are, measured in standard deviations.
Frequently asked
What's the difference between relative and absolute risk?
Relative risk (RR) is the ratio of risks (“doubles your risk”); absolute risk is the actual change in percentage points. A doubling can be 1-in-a-million becoming 2-in-a-million — dramatic relatively, negligible absolutely. Always ask for the absolute numbers.
What is the number needed to treat (NNT)?
How many people must be treated for one extra person to benefit. It's the inverse of the absolute risk reduction, so a small absolute benefit means a large (worse) NNT.
When should I report an odds ratio instead of a risk ratio?
Use the risk ratio for risks from cohort or trial data; reserve the odds ratio for case-control designs (or logistic regression). The odds ratio overstates the risk ratio when the outcome is common, so don't read “OR = 3” as “three times the risk.”
Why can a relative risk reduction sound bigger than it is?
Because it ignores how common the outcome was to begin with. “Cuts risk by 50%” sounds dramatic, but if the risk falls from 2 in 1,000 to 1 in 1,000, the absolute reduction is just 0.1 percentage points and you would need to treat 1,000 people to prevent one event. Always ask for the absolute risk reduction and the number needed to treat alongside any headline relative figure.