Introduction to diagnostic test accuracy network meta-analysis

An important step before clinical intervention selection is the diagnosis of the condition of a patient. Diagnostic tests are commonly used to confirm or exclude a target condition (e.g. a disease). Decision-making rarely relies on a single diagnostic test accuracy (DTA) study; instead, evidence from multiple DTA studies addressing the same question is used.

DTA meta-analysis focuses on evaluating individual tests across separate studies. However, multiple index tests may be available for a target condition, making their comparative accuracy is important for decision-making. Network meta-analysis (NMA) allows for a more integrated and comprehensive evaluation of several diagnostic tests simultaneously in a single model.

Over the past few years, several NMA models have been developed to evaluate the comparative accuracy of multiple diagnostic tests. In this Cochrane Learning Live webinar, the presenters will discuss why traditional NMA methods for interventions are unsuitable for DTA studies and highlight basic principles and assumptions of DTA-NMA models, such as handling multiple thresholds. Finally, the session will explore challenges and opportunities in conducting a DTA-NMA.

The session is of particular interest to review authors who would like to incorporate results from DTA studies comparing multiple diagnostic tests in a review.


Presenter Bios

Dr. Areti-Angeliki Veroniki is a Scientist at the Knowledge Translation Program of St. Michael’s Hospital, Unity Health Toronto, and an Assistant Professor at the University of Toronto in the Institute of Health Policy, Management, and Evaluation. She is a co-Convenor of the Cochrane Statistical Methods Group and co-chair of the Cochrane Methods Executive. Her interests are to optimize the processes of evidence-based medicine and statistical modelling for knowledge synthesis of study findings. Specifically, her research focuses on methods for meta-analysis and network meta-analysis.

Dr. Sofia Tsokani is Statistics Editor in the Methods Support Unit of the Cochrane Central Editorial Team. She is also a post-doctoral researcher at the Laboratory of Hygiene, Social & Preventive Medicine and Medical Statistics of Aristotle University of Thessaloniki in Greece. Her research interests encompass evidence synthesis methods and more specifically statistical modelling for meta-analysis and network meta-analysis of both intervention and diagnostic test accuracy studies.


Sign up

Wednesday, 20 November 2024, 14:00 UTC [check the time in your timezone] SIGN UP HERE

You will need a Cochrane Account to register for this webinar. If you don’t have a Cochrane Account you will be able to register for free on the following page. You will be able to use this account for all future activity. A brief guidance on how to sign up using your Cochrane Account is available here and if you have any problems, please contact support@cochrane.org