[SaFEGen]: A Statistical Framework for efficient Evidence Generation in diagnostics

诊断测试是任何现代医疗系统的关键组成部分：据估计，大约70%的临床决策受到体外诊断(IVDS)的影响。它们既可用于诊断，也可用于排除健康不佳的原因。它们还被用来监测、筛查和评估人们是否存在潜在的健康问题。它们越来越多地允许慢性病患者管理自己的病情。然而，NHS预算中只有不到1%专门用于采用新的和创新的IVD产品，通常需要10年以上的时间和大量的资源才能实现新的诊断测试的广泛采用。尽管存在这些挑战，但诊断学在英国经济中发挥着重要作用，雇佣了超过8000人，在英国，IVD市场每年超过11亿英镑。皇家统计学会(RSS)委托撰写了一份关于诊断测试的独立报告，部分原因是“RSS特别关注，许多新的SARS-CoV-2抗原或抗体诊断测试正在进入市场，用于临床实践和监测，但没有足够的条款对其分析和临床表现进行统计评估。在人们对应用于体外诊断测试评估标准的广泛背景感到担忧的情况下，有必要对一般诊断测试的原则，特别是在大流行中的应用，进行明确的统计学思考”。该报告的关键建议之一是，“对体外诊断的临床表现进行设计良好、动力充足、分析正确的研究，对于该测试的每一种预期用途都是重要的。”在这个项目中，我们的目标是开发一个统计框架，以确保不仅诊断研究得到良好的设计、充分的动力和正确的分析，而且能够最有效地利用开发阶段之间的数据，并利用传统临床试验设计中的前沿进步。SAFEGen项目将通过开发尖端集成统计方法来设计和分析诊断研究，从而实现新诊断测试上市时间的阶段性变化。SAFEGen将通过开发适应性(根据迄今观察到的数据在研究期间修改研究设计)和无缝设计(结合单独的研究以减少时间和提高效率)设计，最大限度地利用数据来节省开发新的诊断测试的时间和资金。采用该框架的效果将是一个更快、更有效的证据生成过程，以开发诊断学，而不会损失严谨性，最终，当高质量的诊断结果更快地到达NHS时，患者会得到更好的结果。

Diagnostic tests are a critical component of any modern healthcare system: it is estimated that approximately 70% of clinical decisions are influenced by the use of in vitro diagnostics (IVDs). They are used both to enable diagnosis and to rule out causes of ill health. They are also used to monitor, screen and assess people for potential health problems. Increasingly, they allow people with chronic disease to manage their own conditions. However, less than 1% of the NHS budget is dedicated to the uptake of new and innovative IVD products, and it typically takes more than 10 years and considerable resources to achieve widespread adoption of a novel diagnostic test. Despite these challenges, diagnostics play a significant role in the UK economy, employing over 8,000 people, and in the UK the IVD market accounts for over £1.1 billion pounds annually. An independent report on diagnostic tests was commissioned by the Royal Statistical Society (RSS) in part since "the RSS has been particularly concerned that many new diagnostic tests for SARS-CoV-2 antigen or antibodies were coming to market for use both in clinical practice and for surveillance without adequate provision for statistical evaluation of their analytical and clinical performance. Against a wider background of concern about standards applied to the evaluation of in vitro diagnostic tests, there was a need for clear statistical thinking on the principles of diagnostic testing in general, and their application in a pandemic in particular". One of the key recommendations of the report was that "undertaking well designed, adequately powered and correctly analysed studies of the clinical performance of an in vitro diagnostic is important for each intended use of the test." In this project we aim to develop a statistical framework that ensures that not only are diagnostic studies well designed, adequately powered and correctly analysed, but also make the most efficient use of data between development stages and takes advantage of cutting-edge advances in conventional clinical trial design.The SaFEGen project will achieve a step change in the time to market of novel diagnostic tests by developing cutting-edge integrated statistical methods for the design and analysis of diagnostic studies from inception to adoption into the NHS. SaFEGen will make best use of data to save time and money in the development of novel diagnostic tests by developing adaptive (modifying a study design during a study based on data observed so far) and seamless (combining separate studies to reduce time and increase efficiency) designs. The effect of the uptake of the framework will be a faster and more efficient evidence generation process for development of diagnostics, with no loss of rigour, and, ultimately, better outcomes for patients when high-quality diagnostics reach the NHS more quickly.