Characterising Uncertainty in the Assessment of Medical Devices and Determining Future Research Needs
探讨了医疗器械评估中证据有限导致的不确定性，分析了学习曲线效应、增量创新、沉没成本等特有因素，并比较了“仅在研究中”或“有条件批准”等决策方案对患者早期获取新技术和降低错误决策风险的作用。
Decisions about the adoption of medical interventions are informed by evidence on their costs and effects. For a range of reasons, evidence relating to medical devices may be limited. The decision to adopt a device early in its life cycle when the evidence base is least mature may impact on the prospects of acquiring further evidence to reduce uncertainties. Equally, rejecting a device will result in no uptake in practice and hence no chance to learn about performance. Decision options such as 'only in research' or 'approval with research' can overcome these issues by allowing patients early access to promising new technologies while limiting the risks associated with making incorrect decisions until more evidence or learning is established. In this paper, we set out the issues relating to uncertainty and the value of research specific to devices: learning curve effects, incremental device innovation, investment and irrecoverable costs, and dynamic pricing. We show the circumstances under which an only in research or approval with research scheme may be an appropriate policy choice. We also consider how the value of additional research might be shared between the manufacturer and health sector to help inform who might reasonably be expected to conduct the research needed. © 2017 The Authors. Health Economics published by John Wiley & Sons, Ltd.