RT Journal Article
SR Electronic
T1 Statistical Methods for Modeling Time-Updated Exposures in Cohort Studies of Chronic Kidney Disease
JF Clinical Journal of the American Society of Nephrology
JO CLIN J AM SOC NEPHROL
FD American Society of Nephrology
SP 1892
OP 1899
DO 10.2215/CJN.00650117
VO 12
IS 11
A1 Xie, Dawei
A1 Yang, Wei
A1 Jepson, Christopher
A1 Roy, Jason
A1 Hsu, Jesse Y.
A1 Shou, Haochang
A1 Anderson, Amanda H.
A1 Landis, J. Richard
A1 Feldman, Harold I.
YR 2017
UL http://cjasn.asnjournals.org/content/12/11/1892.abstract
AB When estimating the effect of an exposure on a time-to-event type of outcome, one can focus on the baseline exposure or the time-updated exposures. Cox regression models can be used in both situations. When time-dependent confounding exists, the Cox model with time-updated covariates may produce biased effect estimates. Marginal structural models, estimated through inverse-probability weighting, were developed to appropriately adjust for time-dependent confounding. We review the concept of time-dependent confounding and illustrate the process of inverse-probability weighting. We fit a marginal structural model to estimate the effect of time-updated systolic BP on the time to renal events such as ESRD in the Chronic Renal Insufficiency Cohort. We compare the Cox regression model and the marginal structural model on several attributes (effects estimated, result interpretation, and assumptions) and give recommendations for when to use each method.