2019 |
RP, Kyle; EEM, Moodie; MB, Klein; M, Abrahamowicz Evaluating Flexible Modeling of Continuous Covariates in Inverse-Weighted Estimators Journal Article American Journal of Epidemiology, 2019. Abstract | Links | BibTeX | Tags: Causal inference, Fractional polynomials, Marginal structural models, Model misspecification, Splines @article{RP2019, title = {Evaluating Flexible Modeling of Continuous Covariates in Inverse-Weighted Estimators}, author = {Kyle RP and Moodie EEM and Klein MB and Abrahamowicz M}, url = {https://pubmed.ncbi.nlm.nih.gov/30649165/}, doi = {10.1093/aje/kwz004}, year = {2019}, date = {2019-06-01}, journal = {American Journal of Epidemiology}, abstract = {Correct specification of the exposure model is essential for unbiased estimation in marginal structural models with inverse-probability-of-treatment weights. However, although flexible modeling is commonplace when estimating effects of continuous covariates in outcome models, its use is less frequent in estimation of inverse probability weights. Using simulations, we assess the accuracy of the treatment effect estimates and covariate balance obtained with different exposure model specifications when the true relationship between a continuous, possibly time-varying covariate Lt and the logit of the probability of exposure is nonlinear. Specifically, we compare 4 approaches to modeling the effect of Lt when estimating inverse probability weights: a linear function, the covariate-balancing propensity score, and 2 easy-to-implement flexible methods that relax the assumption of linearity: cubic regression splines and fractional polynomials. Using data from 2 empirical studies, we compare linear exposure models with flexible exposure models to estimate the effect of sustained virological response to hepatitis C virus treatment on the progression of liver fibrosis. Our simulation results demonstrate that ignoring important nonlinear relationships when fitting the exposure model may provide poorer covariate balance and induce substantial bias in the estimated exposure-outcome associations. Analysts should routinely consider flexible modeling of continuous covariates when estimating inverse-probability-of-treatment weights.}, keywords = {Causal inference, Fractional polynomials, Marginal structural models, Model misspecification, Splines}, pubstate = {published}, tppubtype = {article} } Correct specification of the exposure model is essential for unbiased estimation in marginal structural models with inverse-probability-of-treatment weights. However, although flexible modeling is commonplace when estimating effects of continuous covariates in outcome models, its use is less frequent in estimation of inverse probability weights. Using simulations, we assess the accuracy of the treatment effect estimates and covariate balance obtained with different exposure model specifications when the true relationship between a continuous, possibly time-varying covariate Lt and the logit of the probability of exposure is nonlinear. Specifically, we compare 4 approaches to modeling the effect of Lt when estimating inverse probability weights: a linear function, the covariate-balancing propensity score, and 2 easy-to-implement flexible methods that relax the assumption of linearity: cubic regression splines and fractional polynomials. Using data from 2 empirical studies, we compare linear exposure models with flexible exposure models to estimate the effect of sustained virological response to hepatitis C virus treatment on the progression of liver fibrosis. Our simulation results demonstrate that ignoring important nonlinear relationships when fitting the exposure model may provide poorer covariate balance and induce substantial bias in the estimated exposure-outcome associations. Analysts should routinely consider flexible modeling of continuous covariates when estimating inverse-probability-of-treatment weights. |
Research Papers
2019 |
Evaluating Flexible Modeling of Continuous Covariates in Inverse-Weighted Estimators Journal Article American Journal of Epidemiology, 2019. |