Event

Author credits: Gemma L Clayton, Emily Kawabata, Daniel Major-Smith, Chin Yang Shapland, Tim P Morris, Alice R Carter, Alba Fernández-Sanlés, Maria Carolina Borges, Kate Tilling, Gareth J Griffith, Louise AC Millard, George Davey Smith, Deborah A Lawlor, Rachael A Hughes 

Bias from data missing not at random (MNAR) is a persistent concern in health-related research. A bias analysis quantitatively assesses whether conclusions change under different assumptions about missingness. The dependency between missingness and the outcome is typically modelled using either a selection or pattern-mixture model. Both include non-identifiable bias parameters which govern the magnitude and direction of the bias. Probabilistic bias analysis (PBA) specifies a prior distribution for the bias parameters, explicitly incorporating available information and uncertainty about their true values. A Bayesian PBA combines the prior distribution with the data’s likelihood function whilst a Monte Carlo PBA samples the bias parameters directly from its prior distribution. We investigated fitting a logistic regression when the outcome was missing for the majority of participants and data were MNAR depending on the outcome, exposure and auxiliary variables. Via our data example and simulations, we illustrate two PBAs: fully Bayesian selection model and a Monte Carlo pattern-mixture imputation approach.

Rachael Hughes has a Sir Henry Dale Fellowship jointly funded by the Wellcome Trust and the Royal Society. She is a Senior Research Fellow within the Department of Population Health Sciences, Bristol Medical School, University of Bristol. Rachael's main research interests are longitudinal modelling in life-course epidemiology, analysis in the presence of missing data, bias analyses to hidden sources of bias [such as unmeasured confounding, non-random selection, missing data, and measurement error], and clinical epidemiology of HIV and AIDS in the era of antiretroviral therapy. Rachael is an associate editor for Biometrical Journal.

This seminar will be held via Zoom and in-person.

Location:
Ella Latham Auditorium
Ground Floor, Royal Children's Hospital
50 Flemington Road, Parkville
 

Please see below for the link to join the seminar, which is being recorded.

Join here

Or, go to https://monash.zoom.us/join and enter

Meeting ID: 830 5370 1191

Passcode: 439750