Methodologies for the benefit-risk analysis of medical devices: A systematic review

Introduction : The use of medical devices (MDs) by patients carries both benefits and risks. Regulatory frameworks such as the EU Medical Device Regulation (MDR) and the US Federal Food, Drug, and Cosmetic Act (FD&C Act) mandate the systematic weighing of these benefits and risks through a benefit-risk analysis (BRA). This systematic review aims to identify existing BRA methodologies for MDs, and evaluate their strengths and weaknesses. Methods : A literature search was conducted in PubMed, Scopus, Web of Science, Google Scholar, and Semantic Scholar, covering publications from 2000 onwards using a search string that contained the search terms (1) methodologies AND (2) benefit-risk analysis AND (3) medical device OR in vitro diagnostic medical device. Peer-reviewed publications were included when they described BRA methodologies for MDs. The exclusion criteria included records on unrelated technologies, pharmaceuticals, non-English publications, and insufficient descriptions. The quality assessment was performed using a method proposed by Hawker et al. (2002). The methods and their characteristics were narratively summarised. Each method was assessed for its degree of objectivity and for using numerical calculations on a scale from 0 to 5. Results : The search identified 622 records, with six meeting the inclusion criteria. The included studies described ten BRA methodologies: Multicriteria Decision Analysis (MCDA), Health Outcomes Modelling (HOM), Stated-Choice Surveys (SCS), Composite Net Clinical Outcome (CNCO), Pairwise Comparisons (PC), Complete Profile of Benefits and Risks (CPBR), Net Benefit Score and Benefit-Risk Ratio (NBS & BRR), Quantitative Benefit-Risk Assessment (qBRA), Quantitative Benefit-Risk Determination (QBRD), and the FDA’s Benefit-Risk Framework (BRF). Methods varied in their reliance on numerical calculations and their degree of objectivity, with a significant correlation between both. The average level of objectivity was assessed as medium (2.1/5). Discussion : This review describes the ongoing debate between qualitative and quantitative BRA methods. While qualitative methods are often criticised for being subjective and biased, our findings reveal that quantitative methods, though more objective, still exhibit subjectivity, especially in endpoint identification and relevance assignment. Despite the rise of quantitative methods, often originating from the pharmaceutical industry, qualitative methods are often still used by MD decision-makers. Additionally, several quantitative methods are unsuited for certain difficult-to-quantify risks of MDs. Integrating both qualitative and quantitative elements may offer a more comprehensive BRA framework for MDs.