Applying ‘Technology Assessment’ and ‘Evidence Based Medicine’ Theory to Interventional RadiologyPart 1: Suggestions for the Phased Evaluation of New Procedures

Abstract

AIM: To compare and contrast interventional radiology (IR) clinical and research practices with the technology assessment and evidence-based medicine (EBM) paradigms and make suggestions for the phased evaluation of new IR procedures.MATERIALS AND METHODS: Course literature of the Association of University Radiologists' ‘Basic Technology Assessment for Radiologists’ course and the McMaster University Health Information Research Unit's ‘How to Teach Evidence-Based Medicine 1999’ course were used to identify major publications in each discipline. A computer search was performed to seek other relevant literature. A model of traditional development of IR procedures was developed. Suggestions for the phased evaluation of IR procedures were derived.RESULTS: As in diagnostic radiology, several levels of progressively stronger IR study design can be described and related to EBM ‘levels of evidence’. These range from case reports and case series through case-control and cohort studies to randomized controlled trials (RCTs). The major weakness in the existing IR literature is the predominance of small, uncontrolled, case series. Randomized controlled trials are likely to provide the best possible evidence of effectiveness. They are expensive and randomization is sometimes unethical or impractical. Case-control and cohort studies have been under-utilized. Evidence-based medicine indices of benefit and harm have not yet been applied in IR and may have clinical advantages over traditional statistical methods. A literature search (10 years) using MeSH terms ‘radiology, interventional’ and ‘efficacy’ yielded 30 papers. Combining ‘radiology, interventional’ and ‘evidence-based medicine’ yielded no papers. Comparative searches substituting the term ‘diagnostic imaging’ for ‘radiology, interventional’ yielded 4883 and 62 papers, respectively.CONCLUSION: Principles of technology assessment and EBM can be applied to the investigation of new IR procedures. A tool is needed to simplify the application of EBM analytic methods. Better education in research methods is needed to raise the levels of evidence provided by the bulk of IR research and allow new procedures to be introduced into practice appropriately.Malone, D. E. & MacEneaney, P. M. (2000). Clinical Radiology55, 929–937.