Rare events are a common problem in Cochrane reviews, especially for secondary and safety outcomes. In the presence of rare events, standard meta-analytical models have important limitations and may lead to biased results.
The objective of this web clinic was to provide guidance on handling rare events in a Cochrane review, present the properties of the inverse variance method and explain why this method is problematic when outcome data are rare.
Afroditi explained the advantages of one-stage meta-analysis models over two-stage models and went through the assumptions and the properties of some of these models.
Below you will find the videos from the November 2023 webinar. Recordings from other Methods Support Unit web clinics are available here.
Part 1: Presentation
Part 2: Questions and answers
Presenter bio
Afroditi Kanellopoulou - Afroditi is a Statistics Editor in the Methods Support Unit. She is a biostatistician with a strong interest in cancer prevention and progression, evidence synthesis as well as causal inference methods. She was trained at National and Kapodistrian University of Athens, Greece, where she received a Bachelor’s in Mathematics and a Master’s in Biostatistics. Since 2019, she is a Research Associate and a PhD candidate at the Department of Hygiene and Epidemiology, University of Ioannina, Greece. Her doctoral research is focused on the investigation of the associations between a wide range of non-therapeutic modifiable risk factors and colorectal cancer prognosis.
Additional materials
Download the slides from the webinar [PDF].