Intended for healthcare professionals


Benefits and harms in drug trials . . . and other stories

BMJ 2018; 360 doi: (Published 21 March 2018) Cite this as: BMJ 2018;360:k1216

The rule of three

Randomised trials of new drugs are designed to detect benefits, and they’re always underpowered to detect adverse events. One consequence is that the balance of benefit and harm can’t be properly evaluated at the time of regulatory approval because rare adverse effects are unlikely to be discovered until the drug goes on sale. Large numbers of participants are needed to identify rare adverse drug reactions, and the rule of three provides a useful heuristic. If a new medication causes an adverse reaction in 1 in 100 cases, then 300 subjects must be studied to give a …

View Full Text

Log in

Log in through your institution


* For online subscription