Could phase 3 medicine trials be tagged as pragmatic? A case study: The Salford COPD trial



Randomized clinical trials (RCTs) can be classified as explanatory or pragmatic. Currently, explanatory and pragmatic are considered to be the extremes of a continuum: Many trials have some features of both explanatory and pragmatic RCTs. The Salford Chronic Obstructive Respiratory Disease (COPD) trial was an open-label phase 3 RCT assessing an experimental product (fluticasone furoate–vilanterol) vs usual care. The Salford investigators labelled it as “the world's first phase 3 pragmatic RCT” in COPD patients. The evaluation of the Salford trial by means of the PRECIS-2 tool, yielded a mix of both extremes (explanatory and pragmatic) with several of the 9 domains close to the explanatory extreme and few to the pragmatic one. A number of the features could not be considered as being minimal changes over usual clinical practice. Hence, it would be difficult to accept that the Salford COPD trial was a pragmatic RCT. In addition, all trial participants could have been subject to the Hawthorne effect. The scientific community needs to be rigorous enough when using certain terms related to RCT. It is clear that the Salford COPD trial had particular features—sharing some of explanatory phase 3 RCTs and some of pragmatic RCTs. This, however, is not enough to tag it as a “pragmatic” RCT providing “real-world” data. These words should not be used when referring to prelicensed RCT, unless they really describe how was the trial conducted and the type of data gathered—something that with the current clinical trial regulations will only occur in very rare circumstances.