New group to review statistical evidence on testing

The RSS has launched a new working group to advise on the development of statistical evidence for in vitro diagnostic (IVDs) testing for infectious diseases.

The Working Group on Diagnostic Tests was created to respond to concerns about the lack of basic statistical evidence on clinical and analytical performance of a number of new diagnostic tests that are available. The experts are calling for stronger statistical standards going forward, which they seek to inform. 

The accuracy and performance of diagnostic tests are not currently held to a common statistical standard in either the UK, US or EU. This has been a longstanding problem, but the Covid-19 pandemic has brought it to the fore, as numerous testing plans have been announced without proper reference to the scientific evidence that underpins them.

The group aims to review the statistical evidence needed to assure the performance of new IVDs and propose a set of recommendations for regulators and government, to assist in the development of new standards. Group members hope that these will inform testing for other infectious diseases as well as Covid-19.  

Professor Deborah Ashby, RSS President and co-chair of the working group, said: 'The Covid-19 pandemic has shone a light on gaps in knowledge regarding diagnostic testing and the lack of quality assurance, which is so crucial for controlling the disease. It is essential for public confidence that there is transparency around the evidence base which informs these important policy decisions. 

'As statisticians, we want to help decision-makers by setting out clear statistical criteria that should be used to assure the effectiveness of diagnostic tests.'

Professor Jon Deeks, who also co-chairs the group, added: 'We have been concerned that when it comes to Covid-19 diagnostic tests, basic statistical standards that can underpin evaluation of their performance have been missing. We are engaging with regulators to help change this. A strong evidence base is needed for the performance of medical diagnostic tests - not just for Covid-19 but also for other infectious disease epidemics.'

Read the Working Group's terms of reference.

Any stakeholders with an interest in the output of this review can contact

The Working Group’s advice will be made publicly available in autumn 2020. 

Working Group members

  • Deborah Ashby (chair), director of the School of Public Health and chair in medical statistics and clinical trials at Imperial College London and RSS President.
  • Sheila Bird, honorary professor at the University of Edinburgh and is a Visiting Senior Fellow at the MRC Biostatistics Unit in Cambridge. Also a member of the RSS Covid-19 Task Force. 
  • Jon Deeks (chair), professor of biostatistics at University of Birmingham, and leads the Cochrane Collaboration’s test evaluation activities. Also a member of the RSS Covid-19 Task Force.
  • Stephen Evans, professor of pharmacoepidemiology at the London School of Hygiene and Tropical Medicine. Previously advised the European Commission and European Medicines Agency regarding drug safety. 
  • Rafael Perera, professor of medical statistics and director of the statistics group in the Nuffield Department of Primary Care Health Sciences at University of Oxford and director of research methodologies in the Centre for Evidence Based Medicine (Oxford). 
  • Yemisi Takwoingi, professor of test evaluation and evidence synthesis at the University of Birmingham. Has several scientific and advisory roles, including co-convenor of the Cochrane Screening and Diagnostic Test Methods Group. 
  • Olivia Varley-Winter (RSS staff), who recently set up the RSS Covid-19 Task Force, and was formerly the Society’s policy and research manager (2014-2018). Managed programmes that established the Ada Lovelace Institute at the Nuffield Foundation (2018-2020). 

Representations that RSS members have made on this subject so far

Load more