Worldwide, there are at least 230 million invasive procedures performed annually and most of us will undergo several in our lifetime. There is therefore a need for high-quality evidence to underpin this clinical area. Currently, however, there is no widely accepted definition of an invasive procedure and the terms 'surgery' and 'interventional procedure' are characterised inconsistently. We propose a definition for invasive procedures which addresses the limitations of those currently available. Our definition was developed from an analysis of the 3946 papers from the last decade. A preliminary definition was created based on existing definitions and applied to a variety of papers reporting all types of procedures. This definition was continuously updated and applied iteratively to all articles. The definition has three key components: (1) method of access to the body, (2) instrumentation and (3) requirement for operator skill. It therefore encapsulates all types of invasive procedure regardless of the method of access to the body (incision, natural orifice or percutaneous access), and is relevant whatever the clinical discipline (eg, obstetric, cardiac, dental, interventional cardiology or radiology). Crucially, the definition excludes medicinal products, except where their administration occurs within an invasive procedure (and thereby involves operator skill). The application of a universal definition of an invasive procedure will (1) inform the selection of relevant methods for study design, (2) streamline evidence synthesis and (3) improve research tracking, helping to identify evidence gaps and direct research funds.
Keywords: definitions; evidence synthesis; invasive procedures; research tracking; study design; surgery.
© Author(s) (or their employer(s)) 2019. Re-use permitted under CC BY. Published by BMJ.