Number needed to treat: Caveat emptor

Number needed to treat (NNT)-the inverse of the absolute risk reduction res ulting from an intervention-was introduced as a yardstick to describe the h arm as well as the benefit of therapeutic maneuvers. Analysis using NNT wor ks well when comparing two or more interventions that have their impact ove r the same period of time in similar populations or patients. Under other c onditions, however, analysis based on NNT can produce results that diverge widely from the impact that the interventions can be expected to have on ri sk of death. This can happen either for entire populations or for an indivi dual when comparing NNTs for interventions which have their effects on diff erent subsets of the population or when comparing interventions which have their effects over different periods of time. We demonstrate how this can o ccur by comparing the NNTs and effect of intervention on deaths in a popula tion for automatic implantable cardioverter defibrillators (AICDs), heart t ransplantation, and cholesterol lowering through nutritional intervention w ith plant stanol ester. (C) 2001 Elsevier Science Inc. All rights reserved.