DAG: interpretation difference of TOTAL and DIRECT effect in terms of adjusting Could one explain in simple words how examining the total and direct effect differ in terms of adjusting? How to interpret the findings of these two approaches?
DAG

Minimal sufficient adjustment sets for estimating the TOTAL effect of comorbidity on mortality: age
Minimal sufficient adjustment sets for estimating the DIRECT effect of comorbidity on mortality:
age, physical functioning
 A: The distinction between direct and total effect often concerns the role of mediators. In this DAG, we can try to discern "comorbidity" as the exposure of interest, and mortality as the response or outcome. "Age" fulfills the role of confounder, as it is causal of both comorbidity and mortality. Physical functioning is a mediator because it lies on the causal path from comorbidity to mortality. In any model, we would recommend adjustment for age to estimate the unconfounded effects of interest.
The regression estimate of the direct effect of comorbidity on mortality would adjust for physical functioning and age. The expected effect is an attenuation of the effect of comorbidity on mortality (relative to its total effect). Only the weight of the path leading from comorbidity to mortality is of interest.
The regression estimate of the total effect of comorbidity on mortality would adjust for age only. For the moment, ignore the effect of age-- Here not only the weight of the path leading directly from comorbidity to mortality of interest, but also the path from comorbidity to physical functioning and then to comorbidity is of interest. The entire (indirect) mediated paths have a weight given by the product of their respected paths. Now considering the adjustment for age, age confounds the effect between physical functioning and mortality, so the product from age to PF to mortality is subtracted from that indirect effect implicitly through adjustment.
