News

13 members of the ViCBiostat team made the trip to Newcastle (or attended virtually) in the last week of August for the 43rd Annual Conference of the International Society for Clinical Biostatistics (ISCB43).

There were significant strands of discussion throughout the conference in several of ViCBiostat’s main research areas including challenges with missing data, causal inference and mediation, and contemporary trial design and analysis. ViCBiostat investigators presented new research from each of these areas.

Of particular note, John Carlin was the President’s Invited Speaker for the conference, giving a plenary talk entitled “On the many uses and abuses of regression models”, which described the way in which regression methods are often poorly understood and misused in practice, and outlined a plan for reform of regression teaching to emphasise analytic purpose ahead of technique. The presentation was very well received and led to some discussion on Twitter with other attendees and the broader community.  

Many of the network’s early-mid career researchers and PhD students were also featured in the program, with contributed talks and lightning poster presentations:

Rhys Bowden: Simulating binary variables with two levels of clustering

Ghazaleh Dashti: Handling multivariable missing data in causal mediation analysis with a single mediator

Susan Ellul (virtual): Causal machine learning and use of sample splitting in settings with high dimensional confounding

Kelsey Grantham (virtual): The staircase cluster randomised trial design: a pragmatic alternative to the stepped wedge design

Jessica Kasza: Decaying correlation parameter values obtained from previously analysed cluster randomised trials

Rheanna Mainzer: A comparison of strategies for selecting auxiliary variables for multiple imputation

Melissa Middleton: Evaluation of multiple imputation approaches for case-cohort studies with binary outcomes

Ehsan Rezaei (virtual): The Impact of Iterative Removal of Low information Cluster period Cells from a Stepped Wedge Design

Jiaxin Zhang: Recoverability and estimation of causal effects under typical multivariable missingness mechanisms

On the final day of the event, Katherine Lee contributed to the mini-symposium session “STRengthening Analytical Thinking for Observational Studies (STRATOS) initiative” while Myra McGuinness gave a presentation at the Early Career Biostatistician event entitled “Peer Review Process for Manuscripts”.

After years of virtual events, it was great to catch up with international colleagues and collaborators in person. If you missed the event, the full program and all conference details can be found online at https://www.iscb2022.info