Content area
Full Text
ABSTRACT
The randomized controlled trial (RCT) is not a gold standard: it is a good experimental design in some circumstances, but that's all. Potential shortcomings in the design and implementation of RCTs are often mentioned in passing, yet most researchers consider that RCTs are always superior to all other types of evidence. This paper examines the limitations of RCTs and shows that some types of evidence commonly supposed to be inferior to all RCTs are actually superior to many. This has important consequences for research methodology, for quality of care in clinical medicine, and-especially-for research funding policy. Because every study design may have problems in particular applications, studies should be evaluated by appropriate criteria, and not primarily according to the simplistic RCT/non-RCT dichotomy promoted by some prominent advocates of the evidence-based medicine movement and by the research evaluation guidelines based on its principles.
Whoever is seeking documentation of clinical practice needs to be critical enough to avoid the lure of the gold standard in assessing evidence, so as to not end up . . . with fool's gold.
- A. Rosner (2002)
THE RISE OF EVIDENCE-BASED MEDICINE (EBM) has had a profound effect on the way medical research is carried out and evaluated. The dominance of the randomized controlled trial (RCT) is central to these newer practices and ideas. A quick search of the relevant literature over the last 10 years reveals an almost exclusive use of RCTs in the health literature, while observational studies can barely be found. Behind this trend is a widely held belief that the RCT is the most scientific and rigorous study design available. Any study using the RCT design is considered superior to any study not using this design-and nowhere is this demonstrated more clearly than in EBMs method of evidence evaluation, whose evidence hierarchies feature RCTs above all other forms of evidence. As a consequence of such beliefs and practices, one has much more hope of getting a bad RCT funded than any caliber of, say, observational study, both because of the opinions of funding bodies and, less directly, because of the publication policies of journals. So there are clear incentives for researchers to conduct nothing but RCTs if they can possibly help it.
In recent years...