Pilot trials: to publish or not to publish?


Science works best when it’s open and transparent. Researchers and journal publishers are increasingly encouraged to drive forward initiatives to improve the transparency and reproducibility of scientific results.  This is particularly true in the case of how best to report the findings of clinical trials. These large studies often have direct public health implications. However the size and complexity of many clinical trials can leave room for controversies about the best way to report their results.

One on-going debate is whether to report the results of clinical trial pilot studies. Pilot studies, sometimes called vanguard or feasibility studies, are early phase studies designed to evaluate the design and conduct of a projected full scale clinical trial. Typically conducted separately from the full-scale trials, pilot studies tend to be smaller and shorter in duration than the subsequent trial they provide the basis for. They may also differ from the final completed study in intervention, dose and study population.  This difference between the initial focus of the pilot study and the eventual findings of the clinical trial can be used as an argument for not publishing the results of pilot studies. If the results of pilot studies do not always reflect the ‘true’ effects of an intervention, isn’t publishing them misleading?

In a commentary published this week in Trials, Senior Editor Larry Friedman presents the case in favour of publishing pilot trial results. He argues that all information relating to a specific condition can be of use to the scientific community, especially in the era of meta-analyses. Additionally publishing the results of a pilot trial where the clinical trial does not go ahead, might explain what went wrong and why, potentially extremely useful information to researchers contemplating performing a similar study.

Dr Freidman also argues that we should be careful not to allow commercial concerns to stand in the way of the obligation of scientists to report research findings for the good of the wider scientific community. Pointing out that most clinical trials are sponsored by industry, he notes that “drug companies are rarely interested in providing information of use to their competitors”.

Trials, BMC Public Health and BMC Medical Research Methodology are part of a growing number of peer reviewed journals publishing pilot studies. While care needs to be taken to communicate the limitations of these kinds of studies, as Dr Freidman points out; “Science and medicine advance best when there is easy and rapid sharing of findings”.

View the latest posts on the On Medicine homepage