LindaValeri/CMAverse: Causal Mediation Analysis

CMAverse provides a suite of functions for reproducible causal mediation analysis including DAG visualization, statistical modeling and sensitivity analysis. It implements six causal mediation analysis approaches including the regression-based approach by Valeri et al. (2013) <doi: 10.1037/a0031034> and VanderWeele et al. (2014) <doi: 10.1515/em-2012-0010>, the weighting-based approach by VanderWeele et al. (2014) <doi: 10.1515/em-2012-0010>, the inverse odd-ratio weighting approach by Tchetgen Tchetgen (2013) <doi: 10.1002/sim.5864>, the natural effect model by Vansteelandt et al. (2012) <doi: 10.1515/2161-962X.1014>, the marginal structural model by VanderWeele et al. (2017) <doi: 10.1111/rssb.12194>, and the g-formula approach for a single time point exposure and multiple mediators allowing for time varying confounders by Robins (1986) <doi: 10.1016/0270-0255(86)90088-6>. Moreover, CMAverse conducts sensitivity analysis for unmeasured confounding via the E-value approach by VanderWeele et al. (2017) <doi: 10.7326/M16-2607> and Smith et al. (2019) <doi: 10.1097/EDE.0000000000001064>, and sensitivity analysis for measurement error via regression calibration by Carroll et al. (1995) <doi: 10.1201/9781420010138> and SIMEX by Cook et al. (1994) <doi: 10.2307/2290994> and Küchenhoff et al. (2006) <doi: 10.1111/j.1541-0420.2005.00396.x>. CMAverse also supports causal mediation analysis for a case control study. When the dataset contains missing values, CMAverse can conduct multiple imputations for causal mediation analysis.

install.packages("remotes")
remotes::install_github("LindaValeri/CMAverse")