Management of a condition that has potentially life-threatening consequences may not lend itself effectively to the scrutiny of a randomized clinical trial when an observation or no treatment option is offered as part of the trial. This type of trial often experiences a significant rate of crossover of subjects from no treatment to treatment, and when results are analyzed on an intent-to-treat basis, they may fail to resolve the issue under study. These trials are frequently used as Level 1 medical evidence and the potential impact on clinical decision-making and reimbursement can be quite significant and long-lasting. The authors observed this phenomenon during participation in the Positive Impact of Endovascular Options for Treating Aneurysms Early (PIVOTAL) trial and have observed it in an analysis of the Endovascular Aneurysm Repair 2 (EVAR 2) trial. Possible solutions to mitigate the high crossover effect are offered for consideration. Some clinical conditions dealing with potentially life-threatening problems probably do not lend themselves to be studied in randomized prospective clinical trials containing an observation or no treatment arm.
Published by Elsevier Inc.