Marginal structural models to estimate the joint causal effect of nonrandomized treatments

Even in the absence of unmeasured confounding factors or model misspecifica tion, standard methods for estimating the causal effect of time-varying tre atments on survival are biased when (a) there exists a time-dependent risk factor for survival that also predicts subsequent treatment, and (b) past t reatment history predicts subsequent risk factor level. In contrast, method s based on marginal structural models (MSMs) can provide consistent estimat es of causal effects when unmeasured confounding and model misspecification are absent. MSMs are a new class of causal models whose parameters are est imated using a new class of estimators-inverse-probability-of-treatment wei ghted estimators. We use a marginal structural Cox proportional hazards mod el to estimate the joint effect of zidovudine (AZT) and prophylaxis therapy for Pneumocystis carinii pneumonia on the survival of HIV-positive men in the Multicenter AIDS Cohort Study, an observational study of homosexual men . We obtained an estimated causal mortality rate (hazard) ratio of .67 (con servative 95% confidence interval .46-.98) for AZT and of 1.14 (.79, 1.64) for prophylaxis therapy. These estimates will be consistent for the true ca usal rate ratios when the functional forms chosen for our models are correc t and data have been obtained on all time-independent and time-dependent co variates that predict both subsequent treatment and mortality.