Journal article

A comparison of methods to estimate the survivor average causal effect in the presence of missing data: a simulation study

Myra B McGuinness, Jessica Kasza, Amalia Karahalios, Robyn H Guymer, Robert P Finger, Julie A Simpson

BMC Medical Research Methodology | BMC | Published : 2019

Abstract

BACKGROUND: Attrition due to death and non-attendance are common sources of bias in studies of age-related diseases. A simulation study is presented to compare two methods for estimating the survivor average causal effect (SACE) of a binary exposure (sex-specific dietary iron intake) on a binary outcome (age-related macular degeneration, AMD) in this setting. METHODS: A dataset of 10,000 participants was simulated 1200 times under each scenario with outcome data missing dependent on measured and unmeasured covariates and survival. Scenarios differed by the magnitude and direction of effect of an unmeasured confounder on both survival and the outcome, and whether participants who died followi..

View full abstract

Grants

Awarded by National Health & Medical Research Council of Australia (NHMRC)


Awarded by American Health Assistance Foundation


Awarded by Victorian Centre for Biostatistics (NHMRC: Centre of Research Excellence)


Awarded by National Health and Medical Research Council (NHMRC)


Awarded by NHMRC


Funding Acknowledgements

Melbourne Collaborative Cohort Study recruitment was funded by VicHealth and Cancer Council Victoria. Further Melbourne Collaborative Cohort Study funding: the National Health & Medical Research Council of Australia (NHMRC) Program Grant 209057, Capacity Building Grant 251533 and Enabling Grant 396414. The ophthalmic component was funded by the Ophthalmic Research Institute of Australia; American Health Assistance Foundation (M2008-082), Jack Brockhoff Foundation, John Reid Charitable Trust, Perpetual Trustees. M. McGuinness is funded by the Australian Government Research Training Program Scheme and a studentship courtesy of Victorian Centre for Biostatistics (NHMRC: Centre of Research Excellence grant 1035261). J. Simpson is funded by a National Health and Medical Research Council (NHMRC) Senior Research Fellowship 1104975, and R. Guymer by a NHMRC Principal Research Fellowship 1103013. This work was supported by infrastructure from the Cancer Council of Victoria. CERA receives operational infrastructure support from the Victorian government. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.