Education And Debate


BMJ 1995; 310 doi: (Published 22 April 1995) Cite this as: BMJ 1995;310:1056

The number needed to treat: a clinically useful measure of treatment effect

An authors' error occurred in this article by Richard J Cook and David L Sackett (18 February, pp 452-4). In the table the event rate for moderate hypertension in the control group should be 0.02 and not 0.2, leading to a number needed to treat of 125 [not 13].

View Abstract