ABSTRACT
There are a number of valid statistical entities for the reporting of results from trials comparing different interventions. A recent addition to these is the concept of the number needed to treat (NNT) to prevent one adverse outcome and the number needed to harm (NNH). More traditional approaches convey data as the relative risk reduction (RRR) and the absolute risk reduction (ARR). The authors of interventional (comparative) studies may present the data as any one of these statistical modalities (but preferably as all three) and the modality they choose could have a significant effect on how the data are perceived. To calculate the various modalities the controlled event rate (CER), which is the incidence of the study end-point in the control or placebo group and the experimental event rate (EER), the incidence of study end-point in the experimental (interventional) group, are necessary:
• The RRR can be calculated as the CER minus the EER all divided by the CER.
