Nonrandomized comparisons an introduction

While randomized trials are regarded as the gold standard, they are not always easy to perform and in some cases may be infeasible. The typical trial will take many years from the initial idea to the final publication of results. During this time one must agree the trial design, negotiate and secure support from participating clinicians and funding bodies, submit the trial for scientific and ethical review, recruit, treat, assess and follow-up patients, analyse the results and pass peer review before publication. Each of these steps present (necessary) obstacles of varying difficulty, and consequently the structure and commitment required to see a trial through from beginning to end should never be underestimated.

Where problems concerned with conducting randomized trials are considered insurmountable, some form of comparison is often still necessary. Here, and perhaps in the early stages of drug development when phase II data alone do not justify a randomized trial and more data are needed, there are essentially two options. A comparative control group is necessary, and this can be non-concurrent (historical) or concurrent.

0 0

Post a comment