TY - JOUR
T1 - Estimating inverse probability weights using super learner when weight-model specification is unknown in a marginal structural Cox model context
AU - The BeAMS study group
AU - Karim, Mohammad Ehsanul
AU - Platt, Robert W.
AU - Shirani, A.
AU - Zhao, Y.
AU - Evans, C.
AU - Kingwell, E.
AU - van der Kop, M. L.
AU - Oger, J.
AU - Gustafson, P.
AU - Petkau, J.
AU - Tremlett, H.
N1 - Publisher Copyright:
Copyright © 2017 John Wiley & Sons, Ltd.
PY - 2017/6/15
Y1 - 2017/6/15
N2 - Correct specification of the inverse probability weighting (IPW) model is necessary for consistent inference from a marginal structural Cox model (MSCM). In practical applications, researchers are typically unaware of the true specification of the weight model. Nonetheless, IPWs are commonly estimated using parametric models, such as the main-effects logistic regression model. In practice, assumptions underlying such models may not hold and data-adaptive statistical learning methods may provide an alternative. Many candidate statistical learning approaches are available in the literature. However, the optimal approach for a given dataset is impossible to predict. Super learner (SL) has been proposed as a tool for selecting an optimal learner from a set of candidates using cross-validation. In this study, we evaluate the usefulness of a SL in estimating IPW in four different MSCM simulation scenarios, in which we varied the specification of the true weight model specification (linear and/or additive). Our simulations show that, in the presence of weight model misspecification, with a rich and diverse set of candidate algorithms, SL can generally offer a better alternative to the commonly used statistical learning approaches in terms of MSE as well as the coverage probabilities of the estimated effect in an MSCM. The findings from the simulation studies guided the application of the MSCM in a multiple sclerosis cohort from British Columbia, Canada (1995–2008), to estimate the impact of beta-interferon treatment in delaying disability progression.
AB - Correct specification of the inverse probability weighting (IPW) model is necessary for consistent inference from a marginal structural Cox model (MSCM). In practical applications, researchers are typically unaware of the true specification of the weight model. Nonetheless, IPWs are commonly estimated using parametric models, such as the main-effects logistic regression model. In practice, assumptions underlying such models may not hold and data-adaptive statistical learning methods may provide an alternative. Many candidate statistical learning approaches are available in the literature. However, the optimal approach for a given dataset is impossible to predict. Super learner (SL) has been proposed as a tool for selecting an optimal learner from a set of candidates using cross-validation. In this study, we evaluate the usefulness of a SL in estimating IPW in four different MSCM simulation scenarios, in which we varied the specification of the true weight model specification (linear and/or additive). Our simulations show that, in the presence of weight model misspecification, with a rich and diverse set of candidate algorithms, SL can generally offer a better alternative to the commonly used statistical learning approaches in terms of MSE as well as the coverage probabilities of the estimated effect in an MSCM. The findings from the simulation studies guided the application of the MSCM in a multiple sclerosis cohort from British Columbia, Canada (1995–2008), to estimate the impact of beta-interferon treatment in delaying disability progression.
KW - inverse-probability weighting
KW - marginal structural models
KW - multiple sclerosis
KW - super learner
KW - time-dependent confounding
UR - http://www.scopus.com/inward/record.url?scp=85013409024&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85013409024&partnerID=8YFLogxK
U2 - 10.1002/sim.7266
DO - 10.1002/sim.7266
M3 - Article
C2 - 28219110
AN - SCOPUS:85013409024
SN - 0277-6715
VL - 36
SP - 2032
EP - 2047
JO - Statistics in Medicine
JF - Statistics in Medicine
IS - 13
ER -