Relative risk

Relative risk compares two probabilities: the event rate among people who got a treatment or exposure, divided by the event rate among those who did not. It tells you the proportional change in risk but says nothing about how common the outcome was to begin with. That is its main trap. A treatment that cuts risk from 2 in 1,000 to 1 in 1,000 has a relative risk of 0.5, the same headline number as a drop from 400 in 1,000 to 200 in 1,000, yet the real-world benefit differs enormously. When reading a study, ask whether the reported figure is relative or absolute, and look for the underlying event counts in both groups. Pair every relative risk with the baseline risk before judging whether an effect matters. Check the confidence interval too: if it crosses 1.0, the data are compatible with no effect.