Table 12.

Comparisons among the approaches

Variable	Cox Regression with Baseline Exposure	Cox Regression with Time-Varying Exposure	MSM
Effects estimated	“Long-term” effects	“Short-term” effects	Effects of the whole exposure history
Correctly estimate effects of time-updated exposure when time-dependent confounding is present	N/A	No	Yes
Assumptions	Noninformative censoring, proportional hazard	Noninformative censoring, proportional hazard	Noninformative censoring, proportional hazard, exchangeability, positivity, and correct specification of weight models
Programming	Easy	Easy	Tedious

MSM, marginal structural model; N/A, not applicable.