Behandeleffect bij individuele patiënten voorspellen*

Conflict of interest: The institute of F. Visseren received grants from Merck, ZonMW and the Catharijne Foundation Utrecht and payment for lectures from Merck and Astra Zeneca. Y. de Graaf is editor-in-chief at the NTvG. Financial support for this article: none declared.

• Often only mean treatment Effects of a treatment investigated in a clinical trial are often reported as the mean effects. The effect of a treatment in an individual patient usually differs from this mean effect.

• Prior to initiation of a treatment treatment effect in an individual patient can be predicted on the basis of clinical patient characteristics by using existing risk scores or a new prediction model.

• Based on data from the 'Justification for the use of statins in prevention' (JUPITER) trial, we have developed a prediction model for the primary prevention of cardiovascular disease with rosuvastatin.

• We conclude that the individual treatment effect of rosuvastatin can be reliably predicted in a simpler way based on the absolute risk of cardiovascular disease in a patient and the relative effect of treatment found in the trial.

• In clinical practise, the individual treatment effect that could be expected can be weighed up against any potential disadvantages of that treatment. For this decision-making process, we introduced the concept ‘number willing to treat’, or the ultimate number of patients one would be willing to treat before the desired result is achieved in one patient.