Causal mediation analysis in presence of multiple mediators uncausally related

Abstract

Mediation analysis aims at disentangling the effects of a treatment on an outcome through alternative causal mechanisms and has become a popular practice in biomedical and social science applications. The causal framework based on counterfactuals is currently the standard approach to mediation, with important methodological advances introduced in the literature in the last decade, especially for simple mediation, that is with one mediator at the time. Among a variety of alternative approaches, Imai et al. showed theoretical results and developed an R package to deal with simple mediation as well as with multiple mediation involving multiple mediators conditionally independent given the treatment and baseline covariates. This approach does not allow to consider the often encountered situation in which an unobserved common cause induces a spurious correlation between the mediators. In this context, which we refer to as mediation with uncausally related mediators, we show that, under appropriate hypothesis, the natural direct and joint indirect effects are non-parametrically identifiable. Moreover, we adopt the quasi-Bayesian algorithm developed by Imai et al. and propose a procedure based on the simulation of counterfactual distributions to estimate not only the direct and joint indirect effects but also the indirect effects through individual mediators. We study the properties of the proposed estimators through simulations. As an illustration, we apply our method on a real data set from a large cohort to assess the effect of hormone replacement treatment on breast cancer risk through three mediators, namely dense mammographic area, nondense area and body mass index.

Epigenetic Alterations of Maternal Tobacco Smoking during Pregnancy: A Narrative Review

“…Mediation analyses usually report the mediated effects for each mediator separately, which is not informative regarding the overall mediated effect. In fact, methods considering each potential mediator separately do not allow efficient identification of the indirect effects when mutual influences exist among the mediators [ 174 ]. Therefore, mediation results considering each CpG independently and estimating the indirect effect separately from the other CpGs (or mediators) should be interpreted with caution.…”

Shared Time Scarcity and the Pursuit of Extraordinary Experiences

“…A 5,000-sample bootstrap analysis using PROCESS Model 4 (Hayes, 2018) revealed a significant indirect effect ( ab ) through partner-focused relationship-maintenance motivation ( b = 0.14, SE = 0.07, 95% CI = [0.02, 0.31]), but not through self-focused relationship-maintenance motivation ( b = 0.05, SE = 0.06, 95% CI = [−0.06, 0.17]). Next, we conducted a robustness check using the R package multimediate (Version 2; Jérolon et al, 2021), which yielded the same pattern: The effect of shared time scarcity on preference for extraordinary experiences was mediated by partner-focused ( b = 0.14, 95% CI = [0.02, 0.31]) and not by self-focused ( b = 0.05, 95% CI = [−0.06, 0.17]) relationship-maintenance motivation.…”

Association between menopausal hormone therapy, mammographic density and breast cancer risk: results from the E3N cohort study

“…In order to exclude the possibility that the differences between the results of the partially and completely adjusted mediation models were due the presence of missing values, both analyses were performed after excluding 48 case-control pairs with missing values in any of the potential confounders. Second, to account for the correlation between PMD and BMI in evaluating their joint role as mediators of the effect of MHT on BC risk, we applied a modi ed version of the quasi-Bayesian algorithm by Imai et al 18 described elsewhere 19 . In this model, the linear relationship between the squared root transformed PMD and MHT was estimated using the weights to account for the over-representations of BC cases, as described above.…”