Several strategies are available to address this problem. One approach is to group all ordered mediators into a single mediator of interest and decompose the total effect into effect via or not via this grouped mediator 17, 18. Another approach is to assume a strong hypothesis across the world to measure the upper and lower limits. of PSE by sensitivity analysis, while the point estimate of PSE is still unobtainable 19. Lin and VanderWeele proposed an interventional approach to estimate path-specific effect analogues without unmeasured confounding assumptions with an approach based on regression and a corresponding SAS code 20. However, several limitations of this method are noted. The first is that the result must follow an ordinary linear regression model and cannot be fitted to nonlinear or generalized linear models. Moreover, unlike the overall effect analysis, the analytical solutions for all PSE estimates vary considerably across different models, even though the result follows a linear model. Therefore, software based on has only few models