On this page you can find the content from A day with… Statistical Methods Group which took place on 11 May 2021.

Programme (11 May 2021)

Session 1: Methods and issues around heterogeneity
Session 2: Presentations from early career researchers & Virtual 'get-to-know-you'
Virtual Facewall
The floor is open...

Session 1: Methods and issues around heterogeneity

This was a 60-min session with a series of brief presentations and Q&A. Chaired by Joanne McKenzie and Areti Angeliki (Argie) Veroniki (SMG Co-convenors).

Click on the presentation titles below to watch the recordings.

Introduction Methods and issues around heterogeneity   Download slides: [PDF]	Presentation 1: Extending the I-squared statistic to describe treatment effect heterogeneity in cluster, multi-centre randomized trials and individual patient data meta-analysis   Download slides: [PDF]	Presentation 2: The empirical distribution of tau from IQWiG reports for the application in Bayesian random-effects meta-analyses   Download slides: [PDF]
Julian Higgins University of Bristol, UK	Karla Hemming University of Birmingham, UK	Ralf Bender Institute for Quality and Efficiency in Health Care (IQWiG), Germany
Presentation 3: I2 statistic in meta-analysis of prevalence: worthwhile or worthless?   Download slides: [PDF]	Discussant: Reflections on quantifying heterogeneity   Download slides: [PDF]	Questions and answers
Celina Borges Migliavaca National Institute for Health Technology Assessment, Post-Graduate Program in Epidemiology, Federal University of Rio Grande do Sul, Brazil	Ian White Medical Research Council Clinical Trials Unit, University College London, UK

Session 2: Presentations from early career researchers & Virtual 'get-to-know-you'

This session consisted of two parts. First, there was a 90-min live online session with a series of brief presentations from early career researchers and Q&A, chaired by Anna Chaimani, Catrin Tudur Smith and Mark Simmonds (SMG Co-convenors). This was followed by a 30-min virtual 'get-to-know-you' session, where attendees had a chance to talk about presentations, their own research, getting involved with the Statistical Methods Group, or other topics.

Presentation 1: Meta-analysis methods used in systematic reviews of interrupted time series studies     Download slides: [PDF]	Presentation 2:  Adjusting for exposure misclassification in an individual participant data meta-analysis of observational studies        Download slides: [PDF]	Presentation 3: Latent class models for individual participant data meta-analyses of diagnostic test accuracy studies with imperfect reference standards   Download slides: [PDF]	Presentation 4: Framework for evaluating reporting bias in network meta-analysis   Download slides: [PDF]
Elizabeth Korevaar School of Public Health and Preventive Medicine, Monash University, Australia	Valentijn de Jong Universitair Medisch Centrum (UMC), Utrecht, The Netherlands	Zelalem Negeri Department of Epidemiology, Biostatistics and Occupational Health, McGill University, Canada	Virginia Chiocchia Institute of Social and Preventive Medicine, University of Bern, Switzerland

Presentation 5:  Penalized regression in network meta-analysis: a new approach for analyzing networks of interventions with rare events   Download slides [PDF]	Presentation 6: Using a distribution-based approach and systematic review methods to derive minimum clinically important differences    Download slides [PDF]	Presentation 7:  Meta-analysis for zero-events studies   Download slides [PDF]
Theodoros Evrenoglou Inserm Research Centre in Epidemiology and Statistics (CRESS-U1153), Université de Paris, France	Jennifer Watt Department of Medicine, University of Toronto, Canada	Chang Xu Department of population medicine, Qatar University, Qatar

Virtual Facewall

The Virtual Facewall includes brief videos of Statistical Methods Group co-convenors and presenters talking about themselves, their professional interests and activities.

Visit the Virtual Facewall

The floor is open...

A Twitter chat (#adaywithCochraneSMG) was open during the event, providing opportunities for the community to interact with the Methods Group, raise and discuss methodological and other relevant issues and questions.

If you have any questions about 'A day with... Statistical Methods Group' contact us at support@cochrane.org

Are you interested in joining a Cochrane Methods Group? More info on how to join is available here.