Ending trials early can cause misleading findings; study

25-Mar-2010 - Last updated on 25-Mar-2010 at 11:31 GMT

Clinical trials that finish early due to positive treatment effects often have misleading findings, according to a journal paper which recommends researchers resist pressures to end studies prematurely.

Large, convincing differences between the effect of an experimental treatment and existing therapy can prompt researchers to end a trial, allowing all patients to take the studied drug.

Furthermore, ending a trial early can help a drug reach market sooner and be beneficial for pharma companies, doctors and researchers, according to Victor Montori, an author of the study. However, concerns have been raised that ending trials early could be detrimental to patients.

According to a paper published in the Journal of the American Medical Association (JAMA) truncating a clinical trial can result in misleading estimates of treatment effects. In turn, this can lead to misguided decisions about the trade-off between risks and benefits of a therapy.

Consequently, the paper’s authors, who work at a number of institutions around the globe, recommend that researchers resist ending a trial unless there is a very good reason and the study is almost over anyway.

Ending clinical trials early has been the subject of some debate and controversy but is regarded by the US National Cancer Institute as a valid practice if there are severe side effects or the study “is going particularly well”.

Results were generated by analysing 91 truncated clinical trials asking 63 different questions and 424 matching studies that ran to completion. The analysis showed truncated clinical trials were associated with greater effect sizes, with the impact particularly pronounced in smaller studies.