Abstract
Marginal structural models (MSMs) are commonly used to estimate causal intervention effects in longitudinal nonrandomized studies. A common challenge when using MSMs to analyze observational studies is incomplete confounder data, where a poorly informed analysis method will lead to biased estimates of intervention effects. Despite a number of approaches described in the literature for handling missing data in MSMs, there is little guidance on what works in practice and why. We reviewed existing missing-data methods for MSMs and discussed the plausibility of their underlying assumptions. We also performed realistic simulations to quantify the bias of 5 methods used in practice: complete-case analysis, last observation carried forward, the missingness pattern approach, multiple imputation, and inverse-probability-of-missingness weighting. We considered 3 mechanisms for nonmonotone missing data encountered in research based on electronic health record data. Further illustration of the strengths and limitations of these analysis methods is provided through an application using a cohort of persons with sleep apnea: the research database of the French Observatoire Sommeil de la Fédération de Pneumologie. We recommend careful consideration of 1) the reasons for missingness, 2) whether missingness modifies the existing relationships among observed data, and 3) the scientific context and data source, to inform the choice of the appropriate method(s) for handling partially observed confounders in MSMs.
Highlights
T Abstract RIP Marginal structural models (MSMs) are commonly used to estimate causal intervention effects in C longitudinal non-randomised studies
Despite a number of approaches described in the literature to handle MA missing data in MSMs, there is little guidance on what works in practice and why
We present a simulation study comparing the performance of complete case (CC) analysis, last observation carried forward (LOCF), multiple imputation (MI), inverse-probability of-missingness-weighting (IPMW) and missingness pattern approach (MPA) to handle partially observed confounders under
Summary
We performed a simulation study to (i) illustrate the impact on bias of violations of the assumptions. The bias for the N MPA arises from the direct associations existing between the confounders and the treatment allocation U at subsequent time points even among participants with missing covariate values. When missingness depended on the values of past treatment assignment and confounders, IPMW estimates were unbiased at the three time points. IPT Missingness on constant values R Only LOCF was unbiased (Figure 3) and the bias was worse with MI and IPMW than with CC This is SC because they both use the existing relationships between the confounders, treatment and outcome in U the observed data, but in this scenario, these relationships do not reflect the associations existing MAN between the true (missing) confounder values and the other variables. We would not recommend the implementation of a method known to be biased when unbiased alternatives exist
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
