This site is intended for health professionals only

Be careful about using relative risk

The headline of your recent article on tiotropium is misleading.

News: Tiotropium cuts hospital admissions by a third

It makes the mistake of referring to relative risk reduction and not absolute risk reduction. The absolute risk reduction is the value that we need and the value that most will understand.

Relative risk reductions on their own have little or no meaning in clinical practice. So for instance, when we look at the data in this paper we find that the absolute exacerbation rate in the tiotropium group is 0.64 per year compared with 0.72 per year in the sameterol group – an absolute difference of 0.08 per year.

The widespread use of relative risk reduction in medical journalism is misleading and not to be recommended.

From Dr Mike Ward, Consultant, Sherwood Forest Hospitals NHS Trust

Editor's note

The headline in print avoided a numerical comparison – it simply read ‘Tiotropium reduces COPD admission risk' – but both versions do refer to relative risk reductions.

I agree it would have been useful to have included the absolute risk reduction in the paper too, but the relative risk reductions were reported prominently by the study authors, and are a commonly accepted way of presenting trial results.