How can research questions on the clinical efficacy of an intervention be answered rapidly? Today, the international community is looking for new ways to promptly respond to urgent questions of policy makers and health care providers.
A few agencies (for example the Canadian Agency for Drugs and Technologies in Health) developed tools for carrying out so-called "Rapid Reviews". For the moment, there does not really exist a standardized definition of the concept. When a "condensed" review of the existing scientific knowledge is conducted, the methodology of systematic reviews is mostly used. To save time, certain stages are then simplified and/or skipped.
As a result, the term Rapid Review currently covers a wide range of documents that can be very heterogeneous, both in terms of the scope of the research, as the depth of the analysis.
Like other similar scientific organizations, KCE is constantly looking for the right balance between scientific rigor, efficiency and answering the needs of its customers, users,... In that context, a KCE team conducted a (quick) review of the scientific literature on Rapid Reviews.
The result is a very brief report with all possible 'shortcuts' for the KCE: limiting the scope of the research question and the comprehensiveness (e.g. start from a previously published systematic review,....), easing of the quality control (e.g. no double data collection for all selected studies ...). In any event, each method applied should be described in all transparency. Also for other research teams who want to combine speed and quality, our report may be of interest.
However, a Rapid Review should not evolve into a Discount Review. The strict principles of rigorous systematic reviews should always be respected, and any bias in the selection of the publications, in their evaluation, and in the synthesis of the collected data must be avoided.
Our next appointment is in December 2018, when the results of the Cochrane Group (Cochrane Reviews Rapid Methods Group), which also works on this topic, will be published.