Читать книгу Interventional Cardiology - Группа авторов - Страница 158

Most major trials use analysis by intention to treat whereby all patients are included in their randomized groups, even though they did not all fully comply with the intended treatments. Such an analysis gives an unbiased comparison of the treatment policies as they were delivered in practice, a so‐called pragmatic trial. Per protocol analyses, which exclude any patient when not on randomized treatment, are potentially biased as it could be the sicker patients who opt out. Moreover, many trials reported analyses by modified intention to treat: this method is similar to intention to treat analyses, but allows the exclusion from the analyses of randomized patients for some justified – and possibly prespecified – reasons. For instance, many trials evaluating the safety of novel strategies present the analyses by modified intention to treat. The Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation (ARISTOTLE)[9] and the Effective Anticoagulation with Factor Xa Next Generation in Atrial Fibrillation–Thrombolysis in Myocardial Infarction 48 (ENGAGE‐AF 48) trials [10] – that compared apixaban and edoxaban, respectively, to warfarin in patients with atrial fibrillation – included only bleeding events occurred in randomized patients who received at least one dose of the study drug (modified intention to treat population) in their primary safety analysis, and not all randomized patients.